(N.P. Bali, Manish Goyal) A Textbook of Engineerin PDF

www.cgaspirants.
com
m
co
s.
nt
i ra
sp
ga
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
A TEXTBOOK OF
ENGINEERING MATHEMATICS
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
A TEXTBOOK OF
ENGINEERING
MATHEMATICS
m
For
o
B.TECH. 2nd YEAR, SEMESTER III/IV
.c
Strictly according to the latest revised syllabi (NAS-301/NAS-401)
ts
FOR UTTAR PRADESH TECHNICAL UNIVERSITY (U.P.T.U.), LUCKNOW
n
ra
pi
By
as
N.P. BALI Dr. MANISH GOYAL

Former Principal M.Sc. (Mathematics), Ph.D., CSIR-NET
g
S.B. College, Gurgaon Associate Professor

.c
Haryana Department of Mathematics

Institute of Applied Sciences & Humanities
w
G.L.A. University, Mathura

U.P.
w
w
UNIVERSITY SCIENCE PRESS

(An Imprint of Laxmi Publications Pvt. Ltd.)
An ISO 9001:2008 Company
BENGALURU ● CHENNAI ● COCHIN ● GUWAHATI ● HYDERABAD
JALANDHAR ● KOLKATA ● LUCKNOW ● MUMBAI ● RANCHI ● NEW DELHI
BOSTON (USA) ● ACCRA (GHANA) ● NAIROBI (KENYA)
www.cgaspirants.com
www.cgaspirants.com
A TEXTBOOK OF ENGINEERING MATHEMATICS
© by Laxmi Publications (P) Ltd.

All rights reserved including those of translation into other languages. In accordance with the Copyright (Amendment) Act, 2012,
no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise. Any such act or scanning, uploading, and or electronic sharing of any part of this
book without the permission of the publisher constitutes unlawful piracy and theft of the copyright holder’s intellectual property. If
you would like to use material from the book (other than for review purposes), prior written permission must be obtained from the
publishers.
Printed and bound in India

Typeset at Excellent Graphics, Delhi
First Edition: 2003; Second Edition : 2005; Third Edition: 2006; Fourth Edition: 2010; Revised Fourth Edition : 2010
m
Fifth Edition : 2011; Sixth Edition: 2012; Reprint : 2013, 2014; Seventh Edition: 2015, Eighth Edition : 2016
ISBN 978-93-80386-53-9
o
Limits of Liability/Disclaimer of Warranty: The publisher and the author make no representation or warranties with respect to the
.c
accuracy or completeness of the contents of this work and speciﬁcally disclaim all warranties. The advice, strategies, and activities
contained herein may not be suitable for every situation. In performing activities adult supervision must be sought. Likewise, common
sense and care are essential to the conduct of any and all activities, whether described in this book or otherwise. Neither the publisher
ts
nor the author shall be liable or assumes any responsibility for any injuries or damages arising here from. The fact that an organization
or Website if referred to in this work as a citation and/or a potential source of further information does not mean that the author or
the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers
an
must be aware that the Internet Websites listed in this work may have changed or disappeared between when this work was written
and when it is read.
All trademarks, logos or any other mark such as Vibgyor, USP, Amanda, Golden Bells, Firewall Media, Mercury, Trinity, Laxmi appear-
ir
ing in this work are trademarks and intellectual property owned by or licensed to Laxmi Publications, its subsidiaries or afﬁliates.
Notwithstanding this disclaimer, all other names and marks mentioned in this work are the trade names, trademarks or service
marks of their respective owners.
sp
ga
& Bengaluru 080-26 75 69 30

.c
& Chennai 044-24 34 47 26, 24 35 95 07

& Cochin 0484-237 70 04, 405 13 03
w
& Guwahati 0361-254 36 69, 251 38 81

Branches
& Hyderabad 040-27 55 53 83, 27 55 53 93

w
& Jalandhar 0181-222 12 72

w
& Kolkata 033-22 27 43 84

Published in India by & Lucknow 0522-220 99 16
& Mumbai 022-24 91 54 15, 24 92 78 69
UNIVERSITY SCIENCE PRESS

(An Imprint of Laxmi Publications Pvt.Ltd.)
& Ranchi 0651-220 44 64
An ISO 9001:2008 Company

113, GOLDEN HOUSE, DARYAGANJ,
NEW DELHI - 110002, INDIA
Telephone : 91-11-4353 2500, 4353 2501
Fax : 91-11-2325 2572, 4353 2528 C—
www.laxmipublications.com [email protected] Printed at:
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
Dedicated
pi
To
as
Lord Krishna
g
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
CONTENTS
Preface ... (ix)
Syllabus ... (x)
Standard Results ... (xi)–(xv)

Units Pages
1. Function of Complex Variable ... 1
m
2. Integral Transforms ... 131
3. Statistical Techniques ... 214
o
4. Numerical Techniques–I ... 379
.c
5. Numerical Techniques–II ... 470
Examination Papers ... 555–566
Appendix
n ts ... (i)–(iv)
ra
pi
g as
.c
w
w
w
(vii)
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
PREFACE TO THE EIGHTH EDITION
eÕ;kos'; euks ;s eka fuR;;qDrk miklrsA

J¼;k ij;ksisrkLrs es ;qDrrek erk%AA
The Eighth edition of this textbook is an outcome of new syllabus of Engineering
Mathematics III (NAS-301/NAS-401) for the students of B. Tech. II year (Both III and IV
semester in all branches) proposed and implemented by the U.P. Technical University
(U.P.T.U.), Lucknow recently. The book has been renovated in the light of the latest syllabus.
m
It will work as the latest ready reckoner for the readers.
The subject matter has been made more lucid and easier to understand. A large number
o
of new solved examples and questions have been added. All the answers have been checked
and verified. All the questions of latest university papers have been added in the body of the
.c
text. The suggestions from our colleagues and readers have been incorporated at the proper
places. An appreciably heavy demand of the book ensures its utility to the users.
unit.
n ts
Separate exercise (TEST YOUR KNOWLEDGE) have been given at the end of each
This book is written with a unique style only meant for the welfare of dear students. We
ra
hope that this book will be a strength for them and it will serve their very purpose of attaining
excellent results.
We are highly obliged to Dr. Hari Kishan, Ex-Head, Department of Mathematics,
pi
K.R. (P.G.) College, Mathura, also an eminent author, for his valuable help round the clock in
making this book PERFECT in all senses.
as
We are indebted to GOD for shower of blessing. The suggestions with a view to enhance
the utility of the book are always welcome.
g
—AUTHORS
.c
w
w
w
(ix)
www.cgaspirants.com
www.cgaspirants.com
SYLLABUS
U.P. TECHNICAL UNIVERSITY, LUCKNOW
MATHEMATICS-III
NAS-301/NAS-401 LTP
3 10
Unit I: Function of Complex variable 8
Analytic functions, C-R equations. Harmonic Functions, Cauchy’s integral theorem,
Cauchy’s integral formula. Derivatives of analytic functions, Taylor’s and Laurent’s series,
m
Singularities, Zeroes and Poles, Residue theorem, Evaluation of real integrals of the type
z z +∞
o
2π
f (cos θ, sin θ) dθ and f ( x) dx .
0 −∞
.c
Unit II: Integral Transforms 8
ts
Fourier integral, Complex Fourier transform, Inverse Transforms, Convolution
Theorems, Fourier sine and cosine transform, Applications of Fourier transform to simple one
n
dimensional heat transfer equations, wave equations and Laplace equations, Z-transform and
ra
its application to solve difference equations.
Unit III: Statistical Techniques 8

pi
Moments, Moment generating functions, Skewness, Kurtosis, Curve fitting, Method of

least squares, Fitting of straight lines, Polynomials, Exponential curves, Correlation, Linear,
as
Non-linear and multiple regression analysis, Binomial, Poisson and Normal distributions,
Tests of significations: Chi-square test, t-test.
g
Unit IV: Numerical Techniques–I 8

.c
Zeroes of transcendental and polynomial equations using Bisection method, Regula-

w
falsi method and Newton-Raphson method, Rate of convergence of above methods.

Interpolation: Finite differences, Newton’s forward and backward interpolation,
w
Lagrange’s and Newton’s divided difference formula for unequal intervals.

w
Unit V: Numerical Techniques–II 8

Solution of system of linear equations, Matrix Decomposition methods, Jacobi method,
Gauss Seidal method.
Numerical differentiation, Numerical integration, Trapezoidal rule, Simpson’s one third
and three-eight rules.
Solution of ordinary differential equations (first order, second order and simultaneous)
by Euler’s, Picard’s and fourth-order Runge-Kutta methods.
(x)
www.cgaspirants.com
www.cgaspirants.com
STANDARD RESULTS
d d
1. (xn) = nxn–1 2. (ax) = ax loge a
dx dx
d d 1
3. (ex) = ex 4. (loge x) =
dx dx x
d 1 d
5. (log10 x) = log10 e 6. (sin x) = cos x
dx x dx
d d
7. (cos x) = – sin x 8. (tan x) = sec2 x
m
dx dx
d d
9. (cosec x) = – cosec x cot x 10. (cot x) = – cosec2 x
o
dx dx
.c
d d 1
11. (sec x) = sec x tan x 12. (sin–1 x) =
dx dx 1 − x2
13.
d
dx
(cos–1 x) =
−1
1− x 2
n
14.
d
dx
ts
(tan–1 x) =
1 +
1
x2
ra
d 1 d −1
15. (sec–1 x) = 16. (cot–1 x) =
dx x x2 − 1 dx 1 + x2
pi
d 1 e x − e− x
17. (cosec–1 x) = – 18. sinh x =
2
as
dx x x2 − 1
e x + e− x e x − e− x
19. cosh x = 20. tanh x =
g
2 e x + e− x
.c
21. cosh2 x – sinh2 x = 1, sech2 x + tanh2 x = 1, coth2 x = 1 + cosech2 x

22. cosh2 x + sinh2 x = cosh 2x
w
23. sinh–1 x = log (x + x2 + 1 ), cosh–1 x = log ( x + x2 − 1 )

w
d d
24. (sinh x) = cosh x 25. (cosh x) = sinh x
dx dx
w
d d
26. (tanh x) = sech2 x 27. (coth x) = – cosech2 x
dx dx
d d
28. (sech x) = – sech x tanh x 29. (cosech x) = – cosech x coth x
dx dx
du dv
d dv du d FG u IJ = v dx − u dx
30. Product rule:
dx
(uv) = u
dx
+v
dx
31. Quotient rule:
dx H vK 2
v
dy dy dt
32. = . if y = f1(t) and x = f2(t)
dx dt dx
(xi)
www.cgaspirants.com
www.cgaspirants.com
(xii)
π π π
33. sin–1 x + cos–1 x = , tan–1 x + cot–1 x = , sec–1 x + cosec–1 x =
2 2 2
FG a − b IJ = tan a – tan b, tan F a + b I = tan

34. tan–1
H 1 + ab K
–1
GH 1 − ab JK
–1 –1 –1 a + tan–1 b
F 2x I = sin F 2x I = 2 tan x
35. tan–1 GH 1 − x JK
2 GH 1 + x JK
–1
2
–1
3 tan x − tan3 x
36. sin 3x = 3 sin x – 4 sin3 x, cos 3x = 4 cos3 x – 3 cos x, tan 3x =
1 − 3 tan 2 x
m
2 tan x
sin 2x = 2 sin x cos x, tan 2x = ,
1 − tan 2 x
o
1 − tan 2 x
cos 2x = 2 cos2 x – 1 = 1 – 2 sin2 x = cos2 x – sin2 x =
1 + tan 2 x
.c
x3 x5 x7 x2 x4 x6
37. sin x = x –
ex = 1 + x +
3!
+
5!
x2
+
–
7!
x3
+ ...,
+ ...
n ts
cos x = 1 –
2!
+
4!
–
6!
+ ...
2! 3!
ra
(1 – x)–1 = 1 + x + x2 + x3 + ... ; | x | < 1 (1 + x)–1 = 1 – x + x2 – x3 + ...

(1 – x)–2 = 1 + 2x + 3x2 + 4x3 + ... (1 + x)–2 = 1 – 2x + 3x2 – 4x3 + ...
pi
C+D C−D C+D C−D

38. sin C + sin D = 2 sin cos , sin C – sin D = 2 cos sin
as
2 2 2 2
C+D C−D C+D D−C

cos C + cos D = 2 cos cos , cos C – cos D = 2 sin sin
g
2 2 2 2
.c
39. 2 cos A cos B = cos (A + B) + cos (A – B), 2 sin A sin B = cos (A – B) – cos (A + B)
2 sin A cos B = sin (A + B) + sin (A – B), 2 cos A sin B = sin (A + B) – sin (A – B)
w
40. sin (A + B) = sin A cos B + cos A sin B, sin (A – B) = sin A cos B – cos A sin B
w
cos (A + B) = cos A cos B – sin A sin B, cos (A – B) = cos A cos B + sin A sin B
d 1 d 1
w
41. (sinh–1 x) = , (cosh–1 x) =

dx 1+ x 2 dx 2
x −1
d 1 d 1
(tanh–1 x) = , where | x | < 1, (coth–1 x) = 2 , where | x | > 1
dx 1 − x2 dx x −1
d 1 d 1
(sech–1 x) = – , (cosech–1 x) = –
dx x 1− x 2 dx x x2 + 1
42. (cos θ + i sin θ)n = cos nθ + i sin nθ, (cos θ + i sin θ)–n = cos nθ – i sin nθ
43. sin2 θ + cos2 θ = 1, sec2 θ – tan2 θ = 1, 1 + cot2 θ = cosec2 θ
www.cgaspirants.com
www.cgaspirants.com
(xiii)
44. θ 0° 30° 45° 60° 90° 180° 270° 360°

sin θ 0 1/2 1/ 2 3 /2 1 0 –1 0
cos θ 1 3 /2 1/ 2 1/2 0 –1 0 1
tan θ 0 1/ 3 1 3 ∞ 0 ∞ 0
45. θ 90° – θ 90° + θ π–θ π+θ
sin θ cos θ cos θ sin θ – sin θ
cos θ sin θ – sin θ – cos θ – cos θ
tan θ cot θ – cot θ – tan θ tan θ
m
a b c b2 + c2 − a2
46. sine formula: = = ; cosine formula: cos A =
sin A sin B sin C 2bc
o
a+b+c
47. Area of triangle Δ = s(s − a) (s − b) (s − c) , where s =
.c
2
n!
ts
48. nC =
r r!n− r!
z
n
x n+ 1
49. x n dx = + c; n ≠ – 1
n+1
ra
z 1
x
dx = log x + c;
e z e x dx = ex + c; z a x dx =
ax
log e a
+c
pi
z sin x dx = – cos x + c; z cos x dx = sin x + c

as
z tan x dx = log sec x + c; z cot x dx = log sin x + c

g
z FG π + x IJ + c
.c
sec x dx = log (sec x + tan x) + c = log tan

H 4 2K
z
w
x
cosec x dx = log (cosec x – cot x) + c = log tan +c
2
z z
w
sec x tan x dx = sec x + c; cosec x cot x dx = – cosec x + c

w
50. z a −x2
dx
2
= sin–1
FG x IJ + c;
H aK z − dx
a −x2 2
= cos–1
FG x IJ + c
H aK
z 2
a +x
dx
2
FG x IJ + c; − dx = 1 cot FG x IJ + c
=
H aK
1
a
tan–1
a +x a H aK z 2 2
–1
z dx
2
a −x
=
1
2a 2
F a + x IJ + c; dx = 1 log FG x − a IJ + c
log G
H a − xK x −a 2a H x + aK z 2 2
z dx
x x −a 2
1
= sec G J + c;
a 2
F xI
H a K
− dx
–1
1
= cosec G J + c
a z
F xI
H aK x x 2 − a2
–1
www.cgaspirants.com
www.cgaspirants.com
(xiv)
51. z sech2 x dx = tanh x + c, z cosech2 x dx = – coth x + c
z sinh x dx = cosh x + c, z cosh x dx = sinh x + c
z sech x tanh x dx = – sech x + c, z cosech x coth x dx = – cosech x + c
52. z a2 − x2 dx =
1
2
x a2 − x2 +
1 2
2
x
a sin–1 + c
a
z a2 + x 2 dx =
1
2
x a2 + x2 +
1 2
2
a log (x + a2 + x2 ) + c
m
z x2 − a2 dx =
1
2
x x2 − a2 –
1 2
2
a log (x + x2 − a2 ) + c
o
z FG x IJ + c;
z FG x IJ + c
.c
dx dx
= sinh–1 = cosh–1
a +x2 2 H aK 2
x −a 2 H aK
53. z
a
b
f ( x) dx =
za
b
f ( y) dy ;
za
b
f ( x) dx = –
n
z a
b
f ( x) dx ;
z 0
a
ts
f ( x) dx =
z 0
a
f (a − x) dx
z
R|2 f (x) dx, if f (x) is even functionU|
ra
z
a
f ( x) dx = S V
a
−a |T 0, if f ( x) is odd function |W
0
pi
z 2a
f ( x) dx = S2 z
R| f (x) dx, if f (2a − x) = f (x) U|
a
V
as
T| 0, if f (2 a − x) = − f ( x)W|
0
0
g
54. Leibnitz rule for differentiation under the integral sign
z z
.c
d ψ (α ) ψ (α ) ∂ dψ (α) dφ(α)
f ( x, α) dx = { f ( x, α)} dx + f{ψ(α), α} – f{φ(α), α}
dx φ (α ) φ (α ) ∂α dα dα
w
→
→ → r xi + yj + zk
w
55. If r = xi + yj + zk then | r | = 2

x +y +z 2 2 and r = → =
|r| x 2 + y2 + z 2
w
–→ –→ –→
56. AB = position vector of B-position vector of A = OB – OA
→ → → → → →
57. a ⋅ b = | a | | b | cos θ ; work done = ∫c F ⋅ dr
→ → → →
58. a × b = | a | | b | sin θ n
→ → → →
59. Area of parallelogram = a × b , Moment of force = r × F
www.cgaspirants.com
www.cgaspirants.com
(xv)
→→→
a1 a2 a3 → →
→ → → →
60. a ⋅ ( b × c ) = [ a b c ] = b1 b2 b3 = (a × b) . c
c1 c2 c3
→ → →
where a = Σ a1 i , b = Σ b1 i and c = Σ c1 i
→ → → → → →
If a ⋅ ( b × c ) = 0, then a , b , c are coplanar.
→ → → →
→ → → → → → → → → → → → → a⋅c a ⋅d
61. a × ( b × c ) = (a ⋅ c ) b – ( a ⋅ b ) c 62. ( a × b ) ⋅ ( c × d ) = → → → →
b⋅c b ⋅d
m
→ → → → →→→ → →→→ →
63. ( a × b ) × ( c × d ) = [ a b d ] c – [ a b c ] d
o
64. A (Adj. A) = | A | I 65. AA–1 = I = A–1 A
.c
66. AI = A = IA 67. (ABC)′ = C′B′A′
68. (AB)C = A(BC); A(B + C) = AB + AC
69. A + B = B + A; A + (B + C) = (A + B) + C
n ts
70. (AB)–1 = B–1A–1
ra
71. Walli’s formula

pi
R| n − 1 . n − 3 . n − 5 ...... 3 . 1 . π
z π/2
n
sin θ dθ =
z π/2
cos n
θ dθ = S n n−2 n−4 4 2 2
if n is even
as
0 0 || n . n − 2 . n − 4 ...... 45 . 23
n−1 n−3 n−5
if n is odd
T
g
z eax sin bx dx =
eax
.c
72. (a sin bx – b cos bx) + c

a + b2
2
z
w
eax
eax cos bx dx = (a cos bx + b sin bx) + c
a + b2
2
w
73. Γ(1/ 2) = π , Γ(− 1/ 2) = – 2 π

w
x2 x3 x4 x5 x6
74. log (1 + x) = x – + – + – + ......
2 3 4 5 6
x2 x3 x4 x5 x6
log (1 – x) = – x – – – – – – ......
2 3 4 5 6
75. sin nπ = 0 ; cos nπ = (– 1)n, sin n +

FG 1IJ
π = (– 1)n ; cos n +
1 FG IJ
H 2 K 2 H K
π = 0, where n ∈ I
76. x3 + y3 + z3 – 3xyz = (x + y + z)(x2 + y2 + z2 – xy – yz – zx).
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
UNIT 1
Function of Complex Variable
1.1 INTRODUCTION
m
A complex number z is an ordered pair (x, y) of real numbers and is written as
o
z = x + iy, where i = − 1 .
.c
The real numbers x and y are called the real and Y
imaginary parts of z. In the Argand’s diagram, the complex
P (x, y)
polar coordinates of P, then r =

n
y
ts
number z is represented by the point P(x, y). If (r, θ) are the
x 2 + y 2 is called the modulus
r y
ra
of z and is denoted by | z |. Also θ = tan–1 is called the
x
argument of z and is denoted by arg. z. Every non-zero complex θ
number z can be expressed as
pi
O x M X
z = r (cos θ + i sin θ) = reiθ
If z = x + iy, then the complex number x – iy is called the conjugate of the complex
as
number z and is denoted by z .

Clearly, | z | = | z |,| z |2 = z z ,
g
z+z z−z
.c
Re(z) = , Im(z) = .
2 2i
w
1.2 DEFINITIONS
w
Let S be a non-empty set of complex numbers and δ be a positive real number.

1. Circle. |z – a| = r represents a circle C with centre at the point a and radius r.
w
2. Open disk. The set of points which satisfies the equation |z – z0| < δ defines an open disk
of radius δ with centre at z0 = (x0, y0). This set consists of all points which lie inside circle C.
3. Closed disk. The set of points which satisfies the equation |z – z0| ≤ δ defines a closed disk
of radius δ with centre at z0 = (x0, y0). This set consists of all points which lie inside and on the
boundary of circle C.
4. Annulus. The set of points which lie between two concentric circles C1 : |z – a| = r1 and
C2 : |z – a| = r2 defines an open annulus i.e., the set of points which satisfies the inequality r1
< |z – a| < r2.
The set of points which satisfies the inequality r1 ≤ |z – a| ≤ r2 defines a closed annulus.
It is to be noted that r1 ≤ |z – a| < r2 is neither open nor closed.
www.cgaspirants.com
www.cgaspirants.com
2 A TEXTBOOK OF ENGINEERING MATHEMATICS
5. Neighbourhood. δ-Neighbourhood of a point z0 is the set of all points z for which

|z – z0| < δ where δ is a positive constant. If we exclude the point z0 from the open disk
|z – z0| < δ then it is called the deleted neighbourhood of the point z0 and is written as
0 < |z – z0| < δ.
6. Interior and exterior points. A point z is an interior point of S if all the points in some
δ-neighbourhood of z are in S and an exterior point of S if they are outside S.
7. Boundary point. A point z is a boundary point of S if every δ-neighbourhood of z contains
at least one point of S and at least one point not in S. For example, the points on the circle
|z – z0| = r are the boundary points for the disk |z – z0| ≤ r.
8. Open and closed sets. A set S is open if every point of S is an interior point while a set S
is closed if every boundary point of S belongs to S. e.g. S = {z : |z – z0| < r} is open set while
S = {z : |z – z0| ≤ r} is closed set.
m
9. Bounded set. An open set S is bounded if ∃ a positive real number M such that
| z | ≤ M for all z ∈ S otherwise unbounded.
o
For example: the set S = {z : |z – z0| < r} is a bounded set while the set S = {z : |z – z0| > r} is
.c
an unbounded set.
10. Connected set. An open set S is connected if any two points z1 and z2 belonging to S can
ts
be joined by a polygonal line which is totally contained in S.
11. Domain. An open connected set is called a domain denoted by D.
n
12. Region. A region is a domain together with all, some or none of its boundary points. Thus
ra
a domain is always a region but a region may or may not be a domain.
13. Finite complex plane. The complex plane without the point z = ∞ is called the finite
complex plane.
pi
14. Extended complex plane. The complex plane to which the point z = ∞ has been added is
called the extended complex plane.
as
1.3 FUNCTION OF A COMPLEX VARIABLE

g
If x and y are real variables, then z = x + iy is called a complex variable. If corresponding to

.c
each value of a complex variable z(= x + iy) in a given region R, there correspond one or more
values of another complex variable w (= u + iv), then w is called a function of the complex
w
variable z and is denoted by

w = f(z) = u + iv
w
For example, if w = z2 where z = x + iy and w = f(z) = u + iv

then u + iv = (x + iy) = (x2 – y2) + i(2xy)
2
w
⇒ u = x2 – y2 and v = 2xy
Thus u and v, the real and imaginary parts of w, are functions of the real variables x and y.
∴ w = f(z) = u(x, y) + iv(x, y)
If to each value of z, there corresponds one and only one value of w, then w is called a
single-valued function of z. If to each value of z, there correspond more than one values of w,
then w is called a multi-valued function of z. For example, w = z is a multi-valued function.
To represent w = f(z) graphically, we take two Argand diagrams: one to represent the
point z and the other to represent w. The former diagram is called the XOY-plane or the
z-plane and the latter UOV-plane or the w-plane.
www.cgaspirants.com
www.cgaspirants.com
FUNCTION OF COMPLEX VARIABLE 3
1.4 LIMIT OF f(z)
A function f(z) tends to the limit l as z tends to z0 along any Y

path, if to each positive arbitrary number ε, however small, Z
there corresponds a positive number δ, such that
| f(z) – l | < ε whenever 0 < | z– z0 | < δ
and we write Lt f(z) = l, where l is finite
z → z0
Z0
Note. In real variables, x → x0 implies that x approaches x0 along
the number line, either from left or from right. In complex
variables, z → z0 implies that z approaches z0 along any path,
straight or curved, since the two points representing z and z0 in a
m
O X
complex plane can be joined by an infinite number of curves.
1.5 CONTINUITY OF f(z)
o
.c
A single-valued function f(z) is said to be continuous at a point z = z0 if f(z0) exists, lim f(z)
z → z0
exists and Lt f(z) = f(z0).

z → z0
n ts
A function f(z) is said to be continuous in a region R of the z-plane if it is continuous at
every point of the region. A function f(z) which is not continuous at z0 is said to be discontinuous
ra
at z0.
If the function f(z) = u + iv is continuous at z0 = x0 + iy0 then the real functions u and v
are also continuous at the point (x0, y0). Therefore, we can discuss the continuity of a complex
pi
valued function by studying the continuity of its real and imaginary parts. If f(z) and g(z) are
f ( z)
as
continuous at a point z0 then the functions f(z) ± g(z), f(z) g(z) and , where g(z0) ≠ 0 are also
g( z)
continuous at z0.
g
If f(z) is continuous in a closed region S then it is bounded in S i.e., |f(z)| ≤ M ∀ z ∈ S.
FG IJ
.c
Also, the function f(z) is continuous at z = ∞ if the function f 1 is continuous at ξ = 0

ξ HK
w
1.6 DERIVATIVE OF f(z)

w
Let w = f(z) be a single-valued function of the variable z(= x + iy), then the derivative or
w
differential co-efficient of w = f(z) is defined as
dw f ( z + δz) − f ( z)
= f ′ ( z) = Lt
dz δz → 0 δz
provided the limit exists, independent of the manner in which δz → 0.
1.7 ANALYTIC FUNCTION [G.B.T.U. 2012, M.T.U. 2012, U.P.T.U. 2014]
A function f(z) is said to be analytic at a point z0 if it is one-valued and differentiable not only
at z0 but at every point of some neighbourhood of z0. For example: ex (cos y + i sin y). A function
f(z) is said to be analytic in a certain domain D if it is analytic at every point of D.
www.cgaspirants.com
www.cgaspirants.com
The terms ‘regular’, ‘holomorphic’ and ‘monogenic’ are also sometimes used as synony-
mous with the term analytic.
A function f(z) is said to be analytic at z = ∞ if the function f FG 1IJ is analytic at z = 0.

H zK
Here it should be noted that analyticity implies differentiability but not vice versa. For
example, the function f(z) = |z|2 is differentiable only at z = 0 and nowhere else therefore f(z)
is differentiable at z = 0 but not analytic anywhere.
A function f(z) may be differentiable in a domain except for a finite number of points.
These points are called singular points or singularities of f(z) in that domain.
1.8 ENTIRE FUNCTION
m
A function f(z) which is analytic at every point of the finite complex plane is called an entire
function. Since the derivative of a polynomial exists at every point, a polynomial of any degree
o
is an entire function. Rational functions with non-zero denominators are also entire functions.
.c
1.9 NECESSARY AND SUFFICIENT CONDITIONS FOR f(z) TO BE ANALYTIC
n
The necessary and sufficient conditions for the function
w = f(z) = u(x, y) + iv(x, y)
ts
[M.T.U. 2012, U.P.T.U. (C.O.) 2008]
ra
to be analytic in a region R, are

∂u ∂u ∂v ∂v
(i) , , , are continuous functions of x and y in the region R.
pi
∂x ∂y ∂x ∂y
∂u ∂v ∂u ∂v
(ii) = , =− .
as
∂x ∂y ∂y ∂x
The conditions in (ii) are known as Cauchy-Riemann equations or briefly C-R
equations.
g
Proof. (a) Necessary Condition. Let w = f(z) = u(x, y) + iv(x, y) be analytic in a region R, then
.c
dw
= f ′(z) exists uniquely at every point of that region.
w
dz
Let δx and δy be the increments in x and y respectively. Let δu, δv and δz be the corre-
w
sponding increments in u, v and z respectively. Then,

f ( z + δz) − f ( z) (u + δu) + i(v + δv) − (u + iv)
w
f ′(z) = Lt = Lt
δz → 0 δz δz → 0 δz
FG δu + i δvIJ
= Lt
δz → 0 H δz δz K ...(1)
Since the function w = f(z) is analytic in the region R, the limit (1) must exist independ-
ent of the manner in which δz → 0, i.e., along whichever path δx and δy → 0.
First, let δz → 0 along a line parallel to x-axis so that δy = 0 and δz = δx.
[since z = x + iy, z + δz = (x + δx) + i(y + δy) and δz = δx + iδy]
FG δu + i δv IJ = ∂u + i ∂v
∴ From (1), f ′(z) = Lt
δx → 0 H δx δx K ∂x ∂x ...(2)
Now, let δz → 0 along a line parallel to y-axis so that δx = 0 and δz = iδy.
www.cgaspirants.com
www.cgaspirants.com
f ′(z) = Lt
FG δu + i δv IJ = 1 ∂u + ∂v
∴ From (1),
δy → 0 H i δy i δy K i ∂y ∂y
∂v ∂u 1
= −i ...(3) ∵ =−i
∂y ∂y i
∂u ∂v ∂v ∂u
From (2) and (3), we have +i = −i
∂x ∂x ∂y ∂y
∂u ∂v ∂u ∂v
Equating the real and imaginary parts, = and =− (U.P.T.U. 2015)
∂x ∂y ∂y ∂x
Hence the necessary condition for f(z) to be analytic is that the C-R equations must be
m
satisfied.
(b) Sufficient Condition. Let f(z) = u + iv be a single-valued function possessing partial
o
∂u ∂u ∂v ∂v
derivatives , , , at each point of a region R and satisfying C-R equations.
.c
∂x ∂y ∂x ∂y
∂u ∂v ∂u ∂v
i.e., =
∂x ∂y
and
∂y
=−
∂x
.
ts
We shall show that f (z) is analytic, i.e., f ′(z) exists at every point of the region R.
n
By Taylor’s theorem for functions of two variables, we have, on omitting second and
ra
higher degree terms of δx and δy.

f(z + δz) = u(x + δx, y + δy) + iv(x + δx, y + δy)
LM FG ∂u δx + ∂u δyIJ OP + i LMv(x, y) + FG ∂v δx + ∂v δyIJ OP
pi
= u( x, y) +
N H ∂x ∂y K Q N H ∂x ∂ y K Q
as
= [u(x, y) + iv(x, y)] + G

F ∂u + i ∂vIJ δx + FG ∂u + i ∂vIJ δy
H ∂x ∂x K H ∂y ∂y K
g
.c
= f(z) + G
w
f(z + δz) – f(z) = G

or
w
=G
F ∂u + i ∂v IJ δx + FG − ∂v + i ∂u IJ δy | Using C-R equations
w
H ∂x ∂x K H ∂x ∂x K
F ∂u + i ∂v IJ δx + FG ∂u + i ∂vIJ iδy
= GH
∂x K H ∂x ∂x K |∵ –1=i 2
∂x
=G
F ∂u + i ∂v IJ (δx + iδy) = FG ∂u + i ∂v IJ δz | ∵ δx + iδy = δz
f ( z + δz) − f ( z) ∂u ∂v
⇒ = +i
δz ∂x ∂x
f ( z + δz) − f ( z) ∂u ∂v
∴ f ′(z) = Lt = +i
δz → 0 δz ∂x ∂x
www.cgaspirants.com
www.cgaspirants.com
∂u ∂v
Thus f ′(z) exists, because , exist.
∂x ∂x
Hence f(z) is analytic.
Note 1. The real and imaginary parts of an analytic function are called conjugate functions. Thus, if
f(z) = u(x, y) + iv (x, y) is an analytic function, then u(x, y) and v(x, y) are conjugate functions. The relation
between two conjugate functions is given by C-R equations.
Note 2. When a function f(z) is known to be analytic, it can be differentiated in the ordinary way as if z
is a real variable.
Thus, f(z) = z2 ⇒ f ′(z) = 2z
f(z) = sin z ⇒ f ′(z) = cos z etc.
1.10 CAUCHY-RIEMANN EQUATIONS IN POLAR COORDINATES (U.P.T.U. 2008)
m
Let (r, θ) be the polar coordinates of the point whose cartesian coordinates are (x, y), then
o
x = r cos θ, y = r sin θ,
z = x + iy = r (cos θ + i sin θ) = reiθ
.c
∴ u + iv = f(z) = f(reiθ) ...(1)
Differentiating (1) partially w.r.t. r, we have
∂u
∂r
+i
∂v
∂r
= f ′ (reiθ) . eiθ
n ts ...(2)
Differentiating (1) partially w.r.t. θ, we have
ra
∂u ∂v ∂u FG∂v IJ
∂θ
+i
∂θ
= f ′ (reiθ) . ireiθ = ir
∂r
+i
H∂r K | Using (2)
pi
∂v ∂u
=–r + ir
∂r ∂r
as
Equating real and imaginary parts, we get

∂u ∂v ∂v ∂u
=−r and =r
g
∂θ ∂r ∂θ ∂r
.c
∂u 1 ∂v ∂v 1 ∂u
or = and =− , which is the polar form of C-R equations.
∂r r ∂θ ∂r r ∂θ
w
w
1.11 DERIVATIVE OF w, i.e., f ′(z) IN POLAR COORDINATES
w = f(z)
w
dw ∂u ∂v ∂ ∂w
∴ = f ′ ( z) = +i = (u + iv) =
dz ∂x ∂x ∂x ∂x
∂w ∂r ∂w ∂θ
= +
∂r ∂x ∂θ ∂x
∵ r 2 = x 2 + y2
= cos θ
∂w FG
∂u IJ
∂v sin θ ∴ ∂r/∂x = cos θ as x = r cos θ
∂r
−
H
∂θ
+i
∂θ K
r
and θ = tan −1
yFG IJ
∂w F ∂v ∂u I sin θ x H K
= cos θ − G− r + ir J
∂r H ∂r K r
∂θ − sin θ
∂r ∴ = as y = r sin θ
∂x r
www.cgaspirants.com
www.cgaspirants.com
= cos θ
∂w
−i
∂uFG
+i
∂v IJ sin θ = cos θ
∂w
– i sin θ
∂w
∂r ∂r H ∂r K ∂r ∂r
dw ∂w
⇒ = (cos θ – i sin θ) ...(1)
dz ∂r
∂w
which is the result in terms of .
∂r
dw ∂w ∂r ∂w ∂θ
. .
FG ∂u + i ∂v IJ cos θ − ∂w . sin θ
Again,
dz
= +
∂r ∂x ∂θ ∂x
=
H ∂ r ∂r K ∂θ r
F 1 ∂v − i ∂u IJ cos θ − sin θ ∂w = – i FG ∂u + i ∂v IJ cos θ − sin θ ∂w
= G
m
H r ∂θ r ∂θ K r ∂θ r H ∂θ ∂θ K r ∂θ
o
i ∂w sin θ ∂w
=– cos θ −
r ∂θ r ∂θ
.c
dw i ∂w
⇒ = − (cos θ – i sin θ)
dz r
which is the result in terms of

∂w
.
n ∂θ
ts
∂θ
ra
1.12 HARMONIC FUNCTION [M.T.U. 2014, G.B.T.U. 2012, U.P.T.U. 2007, 2009]
pi
A function of x, y which possesses continuous partial derivatives of the first and second orders
and satisfies Laplace’s equation is called a Harmonic function.
as
1.13 THEOREM
g
If f(z) = u + iv is an analytic function then u and v are both harmonic functions.

.c
Proof. Let f(z) = u + iv be analytic in some region of the z-plane, then u and v satisfy C-R
equations.
w
∂u ∂v
∴ = ...(1)
∂x ∂y
w
∂u ∂v
and =− ...(2)
w
∂y ∂x
Differentiating eqn. (1) partially w.r.t. x and eqn. (2) w.r.t. y, we get
∂ 2u ∂ 2v
= ...(3)
∂x 2 ∂x∂y
∂ 2u ∂ 2v
and = − ...(4)
∂y 2 ∂y∂x
2 2
Assuming ∂ v = ∂ v and adding equations (3) and (4), we get
∂x∂y ∂y∂x
www.cgaspirants.com
www.cgaspirants.com
∂ 2u ∂ 2u
+ =0 ...(5)
∂x 2 ∂y 2
Now, differentiating eqn. (1) partially w.r.t. y and eqn. (2) w.r.t. x, we get
∂ 2u ∂ 2 v
= ...(6)
∂y∂x ∂y 2
∂ 2u ∂2v
and =− 2 ...(7)
∂x∂y ∂x
∂ 2u ∂ 2u
Assuming = and subtracting eqn. (7) from eqn. (6), we get
∂y∂x ∂x∂y
m
∂2v ∂2v
+ =0 ...(8)
∂x 2 ∂y 2
o
Equations (5) and (8) show that the real and imaginary parts u and v of an analytic
function satisfy the Laplace’s equation. Hence u and v are harmonic functions.
.c
Note. Here u and v are called conjugate harmonic functions.
1.14 ORTHOGONAL SYSTEM

n ts [M.T.U. 2012, U.P.T.U. 2009]
Every analytic function f(z) = u + iv defines two families of curves u(x, y) = c1 and v(x, y) = c2,
ra
which form an orthogonal system.
Consider the two families of curves
u(x, y) = c1 ...(1)
pi
and v(x, y) = c2 ...(2)

Differentiating eqn. (1) w.r.t. x, we get
as
∂u Y
∂u ∂u dy dy ∂x
+ . = 0 or =− = m1 (say)
g
∂x ∂y dx dx ∂u
.c
∂y v (x, y) = c2
∂v
w
dy
Similarly, from eqn. (2), we get = − x = m2 (say)
∂
dx ∂v
w
∂y
∂u ∂v
w
. u (x, y) = c1
∂x ∂x
∴ m 1 m2 = ...(3)
∂u ∂v O
. X
∂y ∂y
Since f(z) is analytic, u and v satisfy C-R equations
∂u ∂v ∂u ∂v
i.e., = and =−
∂x ∂y ∂y ∂x
∂v ∂v
.
∂y ∂x
∴ From (3), m1 m2 = =–1
∂v ∂v
− .
∂x ∂y
www.cgaspirants.com
www.cgaspirants.com
Thus, the product of the slopes of the curves (1) and (2) is –1. Hence the curves intersect
at right angles, i.e., they form an orthogonal system.
1.15 THEOREM (U.P.T.U. 2008)
An analytic function with constant modulus is constant.

Proof. Let f(z) = u + iv be an analytic function with constant modulus. Then,
| f(z) | = | u + iv | = constant
⇒ u 2 + v2 = constant = c (say)
Squaring both sides, we get
u 2 + v2 = c 2 ...(1)
m
Differentiating eqn. (1) partially w.r.t. x, we get
∂u ∂v
o
2u + 2v =0
∂x ∂x
.c
∂u ∂v
⇒ u +v =0 ...(2)
∂x ∂x
Again, differentiating eqn. (1) partially w.r.t. y, we get
2u
∂u
∂y
+ 2v
∂v
n
∂y
=0
ts
ra
∂u ∂v
⇒ u +v =0
∂y ∂y
FG IJ FG IJ
pi
∂v ∂u ∂u ∂v ∂v ∂u
and
⇒
∂x
+v
H
u −
∂x
=0
K H K ...(3) ∵
∂y
=−
∂x
=
∂y ∂x
as
Squaring and adding eqns. (2) and (3), we get

R|F ∂u I + F ∂vI U|
2 2
(u2 + v 2) S|GH ∂x JK GH ∂x JK V| = 0
g
T W
.c
FG ∂u IJ + FG ∂v IJ = 0
2 2
u2 + v 2 = c2 ≠ 0
⇒
H ∂x K H ∂x K |∵
w
∂u ∂v
⇒ | f ′(z) |2 = 0 ∵ f ′ ( z) = +i
w
∂x ∂x
⇒ | f ′(z) | = 0
w
⇒ f(z) is constant.
1.16 APPLICATION OF ANALYTIC FUNCTIONS TO FLOW PROBLEMS
Since the real and imaginary parts of an analytic function satisfy the Laplace’s equation in
two variables, these conjugate functions provide solutions to a number of field and flow problems.
For example, consider the two dimensional irrotational motion of an incompressible
fluid, in planes parallel to xy-plane.
→
Let V be the velocity of a fluid particle, then it can be expressed as
→
V = vx i + v y j ...(1)
www.cgaspirants.com
www.cgaspirants.com
Since the motion is irrotational, there exists a scalar function φ(x, y), such that
→ ∂φ ∂φ
i+
V = ∇φ(x, y) = j ...(2)
∂x ∂y
∂φ ∂φ
From (1) and (2), we have vx = and vy = ...(3)
∂x ∂y
The scalar function φ(x, y), which gives the velocity components, is called the velocity
potential function or simply the velocity potential.
→
Also the fluid being incompressible, div V = 0
⇒
FG i ∂ + j ∂ IJ (v i + v j) = 0
H ∂x ∂y K x y
m
∂vx ∂vy
⇒ + =0 ...(4)
∂x ∂y
o
Substituting the values of vx and vy from (3) in (4), we get
.c
FG IJ
∂ ∂φ ∂ ∂φ FG IJ ∂2φ ∂2φ
H K+ = 0 or
H K + =0
n ∂x ∂x ∂y ∂y
ts ∂x 2 ∂y 2
Thus, the function φ is harmonic and can be treated as real part of an analytic function
w = f(z) = φ(x, y) + i ψ (x, y)
ra
For interpretation of conjugate function ψ (x, y), the slope at any point of the curve
ψ (x, y) = c′ is given by
∂φ
pi
∂ψ
dy ∂y
= − ∂x = | By C-R equations
dx ∂ψ ∂φ
as
∂y ∂x
vy
g
= | By (3)
vx
.c
This shows that the resultant velocity vx 2 + vy 2 of the fluid particle is along the tangent
w
to the curve ψ (x, y) = c′ i.e., the fluid particles move along this curve. Such curves are known
as stream lines and ψ (x, y) is called the stream function. The curves represented by
w
φ (x, y) = c are called equipotential lines.

Since φ(x, y) and ψ(x, y) are conjugate functions of analytic function w = f(z), the
w
equipotential lines φ (x, y) = c and the stream lines ψ (x, y) = c′, intersect each other orthogonally.
dw ∂φ ∂ψ ∂φ ∂φ
Now, = +i = −i | By C-R equations
dz ∂x ∂x ∂x ∂y
= vx – ivy | By (3)
dw
∴ The magnitude of resultant velocity = = vx 2 + vy 2
dz
The function w = f(z) which fully represents the flow pattern is called the complex potential.
www.cgaspirants.com
www.cgaspirants.com
In the study of electrostatics and gravitational fields, the curves φ(x, y) = c and ψ (x, y) = c′
are called equipotential lines and lines of force respectively. In heat flow problems, the
curves φ (x, y) = c and ψ (x, y) = c′ are known as isothermals and heat flow lines respectively.
1.17 DETERMINATION OF THE CONJUGATE FUNCTION
If f(z) = u + iv is an analytic function where both u(x, y) and v(x, y) are conjugate functions,
then we determine the other function v when one of these say u is given as follows:
∵ v = v (x, y)
∂v ∂v
∴ dv = dx + dy
∂x ∂y
m
∂u ∂u
⇒ dv = – dx + dy ...(1) | By C-R eqns.
∂y ∂x
o
∂u ∂u
M=– , N=
.c
∂y ∂x
∂M ∂ 2u ∂N ∂ 2 u
ts
∴ = − 2 and =
∂y ∂y ∂x ∂ x 2
∂M ∂N
n
Now, = gives
∂y ∂x
ra
∂ 2u ∂ 2u
– =
∂y 2 ∂x 2
pi
∂ 2u ∂ 2u
or + =0
∂x 2 ∂y 2
as
which is true as u being a harmonic function satisfies Laplace’s equation.

∴ dv is exact.
g
∴ dv can be integrated to get v.

.c
However, if we are to construct f(z) = u + iv when only u is given, we first of all find v by
above procedure and then write f(z) = u + iv.
w
∂v ∂v
Similarly, if we are to determine u and only v is given then we use du = dx − dy
∂y ∂x
w
and integrate it to find u. Consequently f(z) = u + iv can also be determined.

w
1.18 MILNE’S THOMSON METHOD
With the help of this method, we can directly construct f(z) in terms of z without first finding
out v when u is given or u when v is given.
z = x + iy
z = x – iy
1 1
⇒ x = (z + z ) and y = (z – z )
2 2i
∴ f(z) = u(x, y) + iv(x, y)
=u
RS
z+ z z− z
,
UV
+ iv
RS
z+ z z−z
,
UV ...(1)
T2 2i W T 2 2i W
www.cgaspirants.com
www.cgaspirants.com
Relation (1) is an identity in z and z . Putting z = z, we get

f(z) = u(z, 0) + iv(z, 0) ...(2)
Now, f(z) = u + iv
∂u ∂v ∂u ∂u
⇒ f ′(z) = +i = −i | By C-R eqns.
∂x ∂x ∂x ∂y
= φ1(x, y) – i φ2(x, y)
∂u ∂u
where φ 1 ( x, y) = and φ 2 ( x, y) =
∂x ∂y
Now, f ′(z) = φ1 (z, 0) – i φ2(z, 0) | Replacing x by z and y by 0
Integrating, we get
m
f(z) = {φ1(z, 0) – i φ2(z, 0)} dz + c | c is an arbitrary constant.
o
Hence the function is constructed directly in terms of z.
.c
Similarly if v(x, y) is given, then
f(z) = z [ψ1(z, 0) + iψ2(z, 0)] dz + c

n ts ψ 1 ( x, y) =
∂v
∂y
and ψ 2 ( x, y) =
∂v
∂x
Milne’s Thomson method can easily be grasped by going through the steps involved in
ra
following various cases.

Case I. When only real part u(x, y) is given.
pi
To construct analytic function f(z) directly in terms of z when only real part u is given,
we use the following steps:
as
∂u
1. Find
∂x
g
2. Write it as equal to φ1(x, y)

.c
∂u
3. Find
∂y
w

5. Find φ1(z, 0) by replacing x by z and y by 0 in φ1(x, y).
w
6. Find φ2(z, 0) by replacing x by z and y by 0 in φ2(x, y).

7. f(z) is obtained by the formula
w
f(z) =
z {φ 1 ( z, 0) – iφ 2 ( z, 0)} dz + c directly in terms of z.
Case II. When only imaginary part v(x, y) is given.
To construct analytic function f(z) directly in terms of z when only imaginary part v is
given, we use the following steps :
∂v
1. Find
∂y
2. Write it as equal to ψ1(x, y)
∂v
3. Find
∂x
www.cgaspirants.com
www.cgaspirants.com

5. Find ψ1(z, 0) by replacing x by z and y by 0 in ψ1(x, y)
6. Find ψ2(z, 0) by replacing x by z and y by 0 in ψ2(x, y)
7. f(z) is obtained by the formula
f(z) = z
Case III. When u – v is given.
{ ψ 1 ( z, 0) + iψ 2 ( z, 0)} dz + c directly in terms of z.
To construct analytic function f(z) directly in terms of z when u – v is given, we follow

the following steps:
1. f(z) = u + iv ...(1)
2. i f(z) = iu – v ...(2)
m
3. Add (1) and (2) to get
(1 + i) f(z) = (u – v) + i(u + v)
o
or, F(z) = U + iV
.c
where F(z) = (1 + i) f(z), U = u – v and V = u + v
4. Since u – v is given hence U(x, y) is given
5. Find
∂U
∂x
n ts
ra
∂U
7. Find
∂y
pi

9. Find φ1(z, 0)
as
10. Find φ2(z, 0)

11. F(z) is obtained by the formula
z
g
F(z) = { φ1 ( z , 0) – iφ 2 ( z , 0)} dz + c
.c
F( z)
12. f(z) is determined by f(z) = directly in terms of z.
1+ i
w
Case IV. When u + v is given.

w
To construct analytic function f(z) directly in terms of z when u + v is given, we follow

the following steps:
w
1. f(z) = u + iv ...(1)
2. i f(z) = iu – v ...(2)
3. Add (1) and (2) to get
(1 + i) f(z) = (u – v) + i(u + v)
⇒ F(z) = U + iV
where, F(z) = (1 + i) f(z), U = u – v and V = u + v
4. Since u + v is given hence V(x, y) is given
∂V
5. Find
∂y
www.cgaspirants.com
www.cgaspirants.com
∂V
7. Find
∂x
9. Find ψ1(z, 0)
10. Find ψ2(z, 0)
11. F(z) is obtained by the formula
F(z) = z { ψ 1 ( z, 0) + iψ 2 ( z, 0)} dz + c
F( z)
12. f(z) is determined by f(z) = directly in terms of z.
1+ i
m
EXAMPLES
o
Example 1. Find the values of c1 and c2 such that the function
.c
f(z) = x2 + c1y2 – 2xy + i (c2x2 – y2 + 2xy)
is analytic. Also find f ′(z). (U.K.T.U. 2011)
Sol. Here
Comparing (1) with f(z) = u(x, y) + iv(x, y), we get
n
u(x, y) = x2 + c1y2 – 2xy
ts
f(z) = (x2 + c1y2 – 2xy) + i (c2x2 – y2 + 2xy) ...(1)
...(2)
ra
and 2 2
v(x, y) = c2x – y + 2xy ...(3)
For the function f(z) to be analytic, it should satisfy Cauchy-Riemann equations.
∂u
pi
∂u
Now from (2), = 2x – 2y and = 2c1y – 2x
∂x ∂y
∂v ∂v
as
Also, from (3), = 2c2x + 2y and = – 2y + 2x

∂x ∂y
Cauchy-Riemann eqns. are
g
∂u ∂v
.c
=
∂x ∂y
⇒ 2x – 2y = – 2y + 2x which is true.
w
∂u ∂v
and =−
w
∂y ∂x
⇒ 2c1y – 2x = – 2c2x – 2y ...(4)
w
Comparing the coefficients of x and y in eqn. (4), we get

2c1 = – 2 ⇒ c1 = – 1
and – 2 = – 2c2 ⇒ c2 = 1
Hence c1 = – 1 and c2 = 1
∂u ∂v
Now, f ′(z) = +i = 2x – 2y + i(2c2x + 2y)
∂x ∂x
= 2x – 2y + i(2x + 2y) | ∵ c2 = 1
= 2(x + iy) + 2i(x + iy)
= 2z + 2iz = 2(1 + i)z.
Example 2. Find p such that the function f(z) expressed in polar coordinates as
f(z) = r2 cos 2θ + ir2 sin pθ is analytic.
www.cgaspirants.com
www.cgaspirants.com
Sol. Let f(z) = u + iv, then u = r2 cos 2θ, v = r2 sin pθ

∂u ∂v
= 2r cos 2θ, = 2r sin pθ
∂r ∂r
∂u ∂v
= – 2r2 sin 2θ, = pr2 cos pθ
∂θ ∂θ
∂u 1 ∂v ∂v 1 ∂u
For f(z) to be analytic, = and =−
∴ 2r cos 2θ = pr cos pθ and 2r sin pθ = 2r sin 2θ
Both these equations are satisfied if p = 2.
Example 3. (i) Prove that the function sinh z is analytic and find its derivative.
m
(U.K.T.U. 2010)
(ii) Show that f(z) = log z is analytic everywhere in the complex plane except at the origin
o
FG 1IJ .
and that its derivative is
H zK
.c
Sol. (i) Here f(z) = u + iv = sinh z = sinh (x + iy) = sinh x cos y + i cosh x sin y
∴
∂u
n
= cosh x cos y,
∂u
ts
u = sinh x cos y and v = cosh x sin y
= – sinh x sin y
∂x ∂y
ra
∂v ∂v
= sinh x sin y, = cosh x cos y
pi
∂x ∂y
∂u ∂v ∂u ∂v
= =−
as
∴ and
∂x ∂y ∂y ∂x
Thus C-R equations are satisfied.
g
∂u ∂u ∂v ∂v
, ,
.c
Since sinh x, cosh x, sin y and cos y are continuous functions, and are
∂x ∂y ∂x ∂y
also continuous functions satisfying C-R equations.
w
Hence f(z) is analytic everywhere.

w
∂u ∂v
Now f ′(z) = +i = cosh x cos y + i sinh x sin y = cosh (x + iy) = cosh z.
∂x ∂x
w
(ii) Here f(z) = u + iv = log z = log (x + iy)

Let x = r cos θ and y = r sin θ so that
x + iy = r (cos θ + i sin θ) = reiθ
1 FG y IJ
log (x + iy) = log (r eiθ) = log r + iθ =
2
log (x2 + y2) + i tan–1 H xK
Separating real and imaginary parts, we get
1 FG y IJ
u=
2
log (x2 + y2) and v = tan–1
H xK
www.cgaspirants.com
www.cgaspirants.com
∂u x ∂u y
Now, = 2 ,
2 ∂y
= 2
∂x x +y x + y2
∂v −y ∂v x
and = 2 2
, = 2
∂x x +y ∂y x + y2
We observe that the Cauchy-Riemann equations
∂u ∂v ∂u ∂v
= and =–
∂x ∂y ∂y ∂x
are satisfied except when x2 + y2 = 0 i.e., when x = 0, y = 0
Also derivatives are continuous except at origin.
m
Hence the function f(z) = log z is analytic everywhere in the complex plane except at the
origin.
o
∂u ∂v x − iy x − iy 1 1
.c
Also, f ′(z) = +i = 2 = = =
∂x ∂x x + y2 ( x + iy)( x − iy) x + iy z
Example 4. Show that the function ex (cos y + i sin y) is holomorphic and find its
derivative.
Sol.
n
f(z) = ex cos y + i ex sin y = u + iv
ts
Here, u = ex cos y, v = ex sin y
ra
∂u ∂v
= ex cos y = ex sin y
∂x ∂x
pi
∂u ∂v
= – ex sin y = ex cos y
as
∂y ∂y
∂u ∂v ∂u ∂v
Since, = and =−
g
∂x ∂y ∂y ∂x
.c
hence, C-R equations are satisfied. Also first order partial derivatives of u and v are continuous
everywhere. Therefore f(z) is analytic.
w
∂u ∂v
Now, f ′(z) = +i = ex cos y + i ex sin y
w
∂x ∂x
= ex (cos y + i sin y) = ex . eiy = ex+iy = ez
w
Example 5. If n is real, show that rn (cos nθ + i sin nθ) is analytic except possibly when
r = 0 and that its derivative is
nrn–1 [cos (n – 1) θ + i sin (n – 1) θ].
Sol. Let w = f(z) = u + iv = rn (cos nθ + i sin nθ)
Here, u = rn cos nθ, v = rn sin nθ
∂u ∂v
then, = nrn–1 cos nθ = nrn–1 sin nθ
∂r ∂r
∂u ∂v
= – nrn sin nθ = nrn cos nθ
∂θ ∂θ
www.cgaspirants.com
www.cgaspirants.com
∂u 1 ∂v ∂v 1 ∂u
Thus, we see that, = and =−
∴ Cauchy-Riemann equations are satisfied. Also first order partial derivatives of u
and v are continuous everywhere.
dw
Hence f(z) is analytic if f ′(z) or exists for all finite values of z.
dz
dw ∂w
We have, = (cos θ – i sin θ)
dz ∂r
= (cos θ – i sin θ) . nrn–1 (cos nθ + i sin nθ)
= nrn–1 [cos (n – 1) θ + i sin (n – 1) θ]
m
This exists for all finite values of r including zero, except when r = 0 and n ≤ 1.
Example 6. Show that if f(z) is analytic and Re f(z) = constant then f(z) is a constant.
o
(U.P.T.U. 2006)
Sol. Since the function f(z) = u (x, y) + iv (x, y) is analytic, it satisfies the Cauchy-
.c
Riemann equations
Also,
∂u ∂v
=
∂x ∂y
and
n ∂u
∂y
=−
∂v
∂x ts
Re f(z) = constant, therefore u(x, y) = c1
ra
∂u ∂u
∴ =0= .
∂x ∂y
pi
∂v ∂v
Using C-R equations, =0=
∂x ∂y
as
Hence v(x, y) = c2 = a real constant

Therefore f(z) = u(x, y) + iv(x, y) = c1 + ic2 = a complex constant.
g
F I.
.c
y
Example 7. Given that u(x, y) = x2 – y2 and v(x, y) = – GH x 2
+y 2 JK
w
Prove that both u and v are harmonic functions but u + iv is not an analytic function of z.
w
Sol. u = x 2 – y2
∂u ∂ 2u
w
= 2x ⇒ =2
∂x ∂x 2
∂u ∂ 2u
= – 2y ⇒ =–2
∂y ∂y 2
∂ 2u ∂ 2u
Since + =0 Hence u(x, y) is harmonic.
∂x 2 ∂y 2
−y
Also, v=
x + y2
2
∂v 2 xy ∂ 2v 2 y3 − 6 x 2 y
= 2 ⇒ =
∂x (x + y2 )2 ∂x 2 (x 2 + y2 )3
www.cgaspirants.com
www.cgaspirants.com
∂v y2 − x 2 ∂2v 6 x 2 y − 2 y3
= 2 ⇒ 2 =
∂y (x + y2 )2 ∂y (x 2 + y2 )3
∂ 2v ∂ 2v
Since + = 0. Hence v(x, y) is also harmonic.
∂x 2 ∂y 2
∂u ∂v ∂v ∂u
But, ≠ and ≠–
∂x ∂y ∂x ∂y
Therefore u + iv is not an analytic function of z.
Example 8. If φ and ψ are functions of x and y satisfying Laplace’s equation, show that
s + it is analytic, where
m
∂φ ∂ψ ∂φ ∂ψ
s= − and t = + .
∂y ∂x ∂x ∂y
o
[U.K.T.U. 2010, G.B.T.U. (C.O.) 2011]
.c
Sol. Since φ and ψ are functions of x and y satisfying Laplace’s equations,
∂2φ ∂2φ
and
∴ +
∂x 2 ∂y 2
∂2ψ ∂2ψ
+
=0n
= 0.
ts ...(1)
...(2)
∂x 2 ∂y 2
ra
For the function s + it to be analytic,

pi
∂s ∂t
= ...(3)
∂x ∂y
as
∂s ∂t
and =− ...(4)
∂y ∂x
must satisfy.
g
∂s FG
∂ ∂φ ∂ψ IJ
∂2φ ∂2 ψ
.c
= −
Now,
H
∂x ∂x ∂y ∂x
=
K −
∂x∂y ∂x 2
...(5)
w
∂t ∂ F ∂φ ∂ψ I 2 2
= G + J = ∂∂y∂φx + ∂∂yψ
∂y ∂y H ∂x ∂y K 2 ...(6)
w
∂s ∂ F ∂φ ∂ψ I ∂ φ ∂ ψ
2 2
G − J = − ∂y∂x
w
= ...(7)
∂y ∂y H ∂y ∂x K ∂y 2
∂t ∂ F ∂φ ∂ψ I 2 2
and = G + J = ∂∂xφ + ∂∂x∂ψy .
∂x ∂x H ∂x ∂y K 2
...(8)
From (3), (5) and (6), we have

∂2φ ∂2ψ ∂2φ ∂2ψ ∂2ψ ∂2ψ
− = + ⇒ + =0
∂x∂y ∂x 2 ∂y∂x ∂y 2 ∂x 2 ∂y 2
which is true by (2).
www.cgaspirants.com
www.cgaspirants.com
Again from (4), (7) and (8), we have
∂2φ ∂2 ψ ∂2φ ∂2 ψ ∂2φ ∂2φ

− =− 2 − ⇒ + =0
∂y 2
∂y∂x ∂x ∂x∂y ∂x 2 ∂y 2
which is also true by (1).
Hence the function s + it is analytic.
xy 2 (x + iy)
Example 9. Verify if f(z) = , z ≠ 0 ; f(0) = 0 is analytic or not?
x 2 + y4
[U.P.T.U. (C.O.) 2008]
m
xy 2 ( x + iy)
Sol. u + iv = ;z≠0
o
x2 + y4
.c
x 2 y2 xy 3
∴ u= , v =
At the origin,
∂u
= lim
∂x x→0
x2 + y4
x
x2 + y4
u( x, 0) − u(0, 0)
n
= lim
x→0
0−0
x
=0
ts
ra
∂u u (0, y) − u (0, 0) 0−0

= lim = lim =0
∂y y→0 y y → 0 y
pi
∂v v( x, 0) − v(0, 0) 0−0
= lim = lim =0
as
∂x x→0 x x → 0 x
∂v v(0, y) − v(0, 0) 0−0
= lim = lim
g
=0
∂y y→0 y y→ 0 y
.c
∂u ∂v ∂u − ∂v
Since = and =
w
∂x ∂y ∂y ∂x
Hence Cauchy-Riemann equations are satisfied at the origin.
w
f ( z) − f (0) LM
xy 2 ( x + iy) OP 1 xy 2
w
But f ′(0) = lim = lim − 0 . = lim

z→ 0 z x→ 0
y→ 0 N
x2 + y4 Q x + iy x→0 x 2 + y 4
y→ 0
Let z → 0 along the real axis y = 0, then

f ′(0) = 0
Again let z → 0 along the curve x = y2 then
x2 1
f ′(0) = lim 2 2
=
x +x
x→0 2
which shows that f ′(0) does not exist since the limit is not unique along two different paths.
Hence f(z) is not analytic at origin although Cauchy-Riemann equations are satisfied there.
www.cgaspirants.com
www.cgaspirants.com
Example 10. Show that the function defined by f(z) = | xy| is not regular at the origin,
although Cauchy-Riemann equations are satisfied there. [G.B.T.U. (C.O.) 2011]
Sol. Let f(z) = u(x, y) + iv(x, y) = | xy| then u(x, y) = | xy|, v(x, y) = 0
At the origin (0, 0), we have
∂u u( x, 0) − u(0, 0) 0−0
= Lt = Lt =0
∂x x → 0 x x→0 x
∂u u(0, y) − u(0, 0) 0−0
= Lt = Lt =0
∂y y → 0 y y→0 y
∂v v( x, 0) − v(0, 0) 0−0
= Lt = Lt =0
m
∂x x → 0 x x→0 x
∂v v(0, y) − v(0, 0) 0−0
= Lt = Lt =0
o
∂y y → 0 y y→0 y
.c
∂u ∂v ∂u ∂v
Clearly, = , =−
∂x ∂y ∂y ∂x
Hence C-R equations are satisfied at the origin.
Now f ′(0) = Lt
z→0
f ( z) − f (0)
n
z
= Lt
z→0
| xy| − 0
x + iy
ts
ra
If z → 0 along the line y = mx, we get
|mx 2 | |m|
pi
f ′(0) = Lt = Lt
x → 0 x(1 + im) x → 0 1 + im
as
Now this limit is not unique since it depends on m. Therefore, f ′(0) does not exist.
Hence the function f(z) is not regular at the origin.
Example 11. Prove that the function f(z) defined by
g
x 3 (1 + i) − y 3 (1 − i)
.c
f(z) = , z ≠ 0 and f(0) = 0

x2 + y2
w
is continuous and the Cauchy-Riemann equations are satisfied at the origin, yet f ′(0) does not
exist. (U.P.T.U. 2015)
w
( x 3 − y 3 ) + i( x 3 + y 3 )
Sol. Here, f(z) = ,z≠0
w
x2 + y2
x 3 − y3 x 3 + y3
Let f(z) = u + iv = +i ,
x2 + y2 x 2 + y2
x 3 − y3 x 3 + y3
then u= , v =
x 2 + y2 x 2 + y2
Since z ≠ 0 ⇒ x ≠ 0, y ≠ 0
∴ u and v are rational functions of x and y with non-zero denominators. Thus, u, v and
hence f(z) are continuous functions when z ≠ 0. To test them for continuity at z = 0, on changing
u, v to polar co-ordinates by putting x = r cos θ, y = r sin θ, we get
www.cgaspirants.com
www.cgaspirants.com
u = r(cos3 θ – sin3 θ) and v = r (cos3 θ + sin3 θ)

When z → 0, r → 0
∴ Lt u = Lt r (cos3 θ – sin3 θ) = 0
z→0 r→0
Similarly, Lt v = 0
z→0
∴ Lt f(z) = 0 = f(0)
z→0
⇒ f(z) is continuous at z = 0.
Hence f(z) is continuous for all values of z.
At the origin (0, 0), we have
m
∂u u( x, 0) − u(0, 0) x−0
= Lt = Lt =1
∂x x → 0 x x→0 x
o
∂u u(0, y) − u(0, 0) − y−0
= Lt = Lt =–1
∂y y → 0 y y
.c
y→0
∂v v( x, 0) − v(0, 0) x−0
= Lt = Lt =1
∂x x → 0
∂v
= Lt
∂y y → 0
n
y
x
v(0, y) − v(0, 0)
x→0
= Lt
y→0
x
y−0
y
=1
ts
ra
∂u ∂v ∂u ∂v
∴ = and =−
∂x ∂y ∂y ∂x
pi
Hence C-R equations are satisfied at the origin.

f ( z) − f (0) ( x 3 − y 3 ) + i( x 3 + y 3 ) − 0
Now f ′(0) = Lt = Lt
as
z→0 z z→0 ( x 2 + y 2 )( x + iy)

Let z → 0 along the line y = x, then
g
0 + 2ix 3 i i(1 − i) 1 + i
.c
f ′(0) = Lt 3
= = = ...(1)
x→0 2 x (1 + i) 1 + i 2 2
w
Also, let z → 0 along the x-axis (i.e. y = 0), then

x 3 + ix 3
w
f ′(0) = Lt =1+i ...(2)

x→0 x3
Since the limits (1) and (2) are different, f ′(0) does not exist.
w
Example 12. Show that the function f (z) = e − z −4 , z ≠ 0 and f(0) = 0 is not analytic at
z = 0, although Cauchy-Riemann equations are satisfied at this point. [U.P.T.U. (C.O.) 2008]
−4 −4
−z − ( x + iy)
Sol. Here, f(z) = e =e
1
.
( x − iy) 4 −
R|S ( x − iy )
4 U|V
−
( x + iy) ( x − iy) 4
4 |T ( x + y )
2 2 4
|W
=e = e
1
− 2 2 4 [( x 4 + y 4 − 6 x 2 y 2 ) − 4 ixy( x 2 − y 2 )]
(x + y )
=e
www.cgaspirants.com
www.cgaspirants.com
LM x OP
PQ LMcos 4 xy ( x 2 − y 2 ) + i sin 4 xy ( x 2 − y 2 ) OP
4
+ y 4 − 6 x 2 y2
−
⇒ u + iv = e
MN ( x 2 + y2 ) 4
N ( x2 + y2 )4 ( x 2 + y2 ) 4 Q
Lx
−M
4
+ y4 − 6 x2 y2 OP
∴
M
u= e N
( x 2 + y2 ) 4 PQ cos 4 xy ( x 2 − y 2 )
(x2 + y2 )4
−
LM x 4
+ y4 − 6 x 2 y2 OP
and v= e
MN ( x 2 + y2 )4 PQ sin 4 xy ( x 2 − y 2 )
(x2 + y2 )4
∂u u ( x, 0) − u (0, 0)
At z = 0, = lim
∂x x → 0 x
m
−4
e− x −0 1
= lim = lim
o
−4
x→0 x x→0 xe x
.c
1 1
= lim
x→0 L 1
x M1 +
1 OP = lim x + 1
x→0 1
=0
∂u
n N x
u(0, y) − u(0, 0) − e− y
4
+
2x8
+
ts Q
−4
x 3
+
2 x7
+
= lim = lim =0
ra
∂y y→0 y y→0 y
∂v v( x, 0) − v(0, 0) 0
= lim = lim =0
pi
∂x x → 0 x x → 0 x
∂v v(0, y) − v(0, 0) 0
as
and = lim = lim = 0.

∂y y → 0 y y → 0 y
Hence Cauchy-Riemann Conditions are satisfied at z = 0.
g
−4
f ( z) − f (0) e− z
.c
But f ′(0) = lim = lim

z→0 z z→0 z
iπ / 4 −4
w
e − (re )
= lim ; if z → 0 along z = reiπ/4
r→0 re iπ / 4
w
−4
er
= lim=∞
re iπ / 4
w
r→0
which shows that f ′(z) does not exist at z = 0. Hence f(z) is not analytic at z = 0.
Example 13. (i) Examine the nature of the function
R| x y (x + iy) , z ≠ 0U|
2 5
f ( z) = S x + y
4 10 V| in the region including the origin.
|T0 , z = 0W
3R| x y(y − ix) , z ≠ 0U| f(z) − f(0)
(ii) If f(z) = S x + y6 2 V , prove that → 0 as z → 0 along any radius
|T 0 , z = 0 |W
z
vector but not as z → 0 in any manner and also that f(z) is not analytic at z = 0.
[G.B.T.U. 2013, U.K.T.U. 2010]
www.cgaspirants.com
www.cgaspirants.com
x 2 y 5 ( x + iy)
Sol. (i) Here, u + iv = ;z≠0
x 4 + y 10
x 3 y5 x 2 y6
∴ u= 4 10 , v =
x +y x + y 10
4
∂u u( x, 0) − u(0, 0) 0 −0
At the origin, = lim = lim =0
∂x x → 0 x x→0 x
∂u u(0, y) − u(0, 0) 0 −0
= lim = lim =0
∂y y → 0 y y→0 y
∂v ∂v
m
Similarly, =0=
∂x ∂y
∂u ∂v ∂u ∂v
o
Since = and =−
∂x ∂y ∂y ∂x
.c
Hence Cauchy-Riemann equations are satisfied at the origin
LM
x 2 y 5 ( x + iy) OP
But f ′(0) = lim
z→0
n f ( z) − f (0)
x y
z
2 5
= lim
x→0
ts
4
x +y
y→0 N 10
−0 .
Q
1
x + iy
= xlim
ra
→0 x 4 + y 10
y→0
Let z → 0 along the radius vector y = mx, then
pi
m5 x 7 m5 x 3
f ′(0) = lim = lim =0
x → 0 x 4 + m 10 x 10 x → 0 1 + m 10 x 6
as
Again let z → 0 along the curve y5 = x2

1 x4
g
f ′(0) = xlim 4 4
=
x +x
→0 2
.c
which shows that f ′(0) does not exist. Hence f(z) is not analytic at origin although Cauchy-
Riemann equations are satisfied there.
w
(ii)
f ( z) − f (0)
=
LM
x 3 y( y − ix)
− 0 .
1
=
OP
− ix 3 y( x + iy)
.
1
= – i
x3 y
w
z N
x 6 + y2 x + iy Q(x6 + y2 ) x + iy x 6 + y2
Let z → 0 along radius vector y = mx then,
w
f ( z) − f (0) − ix 3 (mx) − imx 2

lim = lim 6 = lim 4 =0
z→0 z x → 0 x + m2 x 2 x → 0 x + m2
f ( z) − f (0)
Hence → 0 as z → 0 along any radius vector.
z
Now let z → 0 along a curve y = x3 then,
f ( z) − f (0) − ix 3 . x 3 − i
lim = lim 6 =
z→0 z x → 0 x + x6 2
f ( z) − f (0)
Hence does not tend to zero as z → 0 along the curve y = x3.
z
We observe that f ′(0) does not exist hence f(z) is not analytic at z = 0.
www.cgaspirants.com
www.cgaspirants.com
Example 14. Show that the following functions are harmonic and find their harmonic
conjugate functions.
1
(i) u = log(x2 + y2) (U.P.T.U. 2015) (ii) v = sinh x cos y.
2
1
Sol. (i) u= log (x2 + y2) ...(1)
2
∂u 1 1 x
= . 2 2
. 2x = 2
∂x 2 x + y x + y2
∂ 2u ( x 2 + y2 ) . 1 − x . 2 x y2 − x2
= = ...(2)
∂x 2 (x 2 + y2 )2 (x2 + y2 )2
m
∂u 1 1 y
Also, = . 2 . 2y = 2
o
2
∂y 2 x + y x + y2
.c
∂ 2u ( x 2 + y2 ) . 1 − y . 2 y x 2 − y2
= = ...(3)
∂y 2 (x2 + y2 )2 (x2 + y2 )2
∂ 2u ∂ 2u
+
∂x 2 ∂y 2
= 0.
n ts [From (2) and (3)]
ra
Since u satisfies Laplace’s equation hence u is a harmonic function.
∂v ∂v
Let dv = dx + dy
pi
∂x ∂y
FG ∂u IJ dx + FG ∂u IJ dy
H ∂y K H ∂x K
= − [Using C-R equations]
as
F − y I dx + F x I dy
g
=G
H x + y JK GH x + y JK
2 2 2 2
.c
x dy − y dx L F yI O
= d Mtan G J P
−1
N H xKQ
w
= 2 2
(x + y )
F yI
w
v = tan G J + c
Integration yields,
H xK
–1 | c is a constant
w
which is the required harmonic conjugate function of u.

(ii) v = sinh x cos y ...(1)
2
∂v ∂ v
= cosh x cos y ⇒ = sinh x cos y ...(2)
∂x ∂x 2
∂v ∂2v
= – sinh x sin y ⇒ = – sinh x cos y ...(3)
∂y ∂y 2
∂2v ∂2v
Since, + =0
∂x 2 ∂y 2
Hence v is harmonic.
www.cgaspirants.com
www.cgaspirants.com
∂u ∂u ∂v ∂v
Now, du = dx + dy = dx − dy
∂x ∂y ∂y ∂x
= – sinh x sin y dx – cosh x cos y dy
= – [sinh x sin y dx + cosh x cos y dy]
= – d (cosh x sin y).
Integration yields, u = – cosh x sin y + c | c is a constant
4 2 2 4
Example 15. (i) Show that the function u(x, y) = x – 6x y + y is harmonic. Also find
the analytic function f(z) = u(x, y) + iv(x, y). (U.P.T.U. 2007)
(ii) Show that the function u = x3 – 3xy2 is harmonic and find the corresponding analytic
function. [U.P.T.U. (C.O.) 2008]
m
x
(iii) Show that e cos y is a harmonic function, find the analytic function of which it is real
part. [U.P.T.U. (C.O.) 2008]
o
Sol. (i) u = x4 – 6x2y2 + y4
.c
∂u ∂ 2u
∴ = 4x3 – 12xy2 ⇒ = 12x2 – 12y2
∂x ∂x 2
∂u
∂y
= – 12 x2y + 4y3 ⇒
n ts
∂ 2u
∂y 2
= – 12x2 + 12y2
ra
∂ 2u ∂ 2u
Since, + =0 ∴ u(x, y) is a harmonic function.
∂x 2 ∂y 2
pi
Now, let dv =
∂v
dx +
∂v FG
dy = −
∂u
dx +
IJ
∂u
dy | By C-R eqns.
∂x ∂y H∂y K
∂x
as
= (12x2y – 4y3) dx + (4x3 – 12xy2) dy

= (12x2y dx + 4x3dy) – (4y3 dx + 12xy2 dy)
g
= d(4x3y) – d(4xy3)
.c
Integration yields, v = 4x3y – 4xy3 + c

w
Hence f(z) = u + iv = x4 – 6x2y2 + y4 + i(4x3y – 4xy3 + c)

= (x + iy)4 + c1 = z4 + c1 | where c1 = ic
w
(ii) u = x3 – 3xy2
∂u ∂ 2u
w
∴ = 3x2 – 3y2 ⇒ = 6x
∂x ∂x 2
∂u ∂ 2u
= – 6xy ⇒ = – 6x
∂y ∂y 2
∂ 2u ∂ 2u
Since, + =0 ∴ u is a harmonic function.
∂x 2 ∂y 2
Now, dv =
dv
dx +
∂v FG IJ
dy = − ∂u dx + ∂u dy | By C-R eqns.
dx ∂y H∂y K
∂x
www.cgaspirants.com
www.cgaspirants.com
= 6xy dx + (3x2 – 3y2) dy = (6xy dx + 3x2 dy) – 3y2 dy

= d(3x2y) – d(y3)
Integration yields,
v = 3x2y – y3 + c
∴ f(z) = u + iv = x3 – 3xy2 + i(3x2y – y3 + c)
= (x + iy)3 + ic = z3 + c1 (where c1 = ic)
(iii) Let u= ex cos y
∂u ∂ 2u
∴ = ex cos y ⇒ = ex cos y
∂x ∂x 2
m
∂u ∂ 2u
= – ex sin y ⇒ = – ex cos y
∂y
o
∂y 2
.c
∂ 2u ∂ 2u
Since + =0 ∴ u is a harmonic function.
∂x 2 ∂y 2
Let dv =
∂v
n
∂x
dx +
∂v
∂y
dy = −
∂u
∂y
FG
dx +
H
∂u
∂x
dy
tsIJ
K
FG IJ
H K | By C-R eqns.
ra
= e x sin y dx + ex cos y dy = d (e x sin y)

Integration yields,
pi
v = e x sin y + c
Hence f(z) = u + iv = e x cos y + i (e x sin y + c)
as
= e x(cos y + i sin y) + c1 | where c1 = ic

= ex + iy + c1 = ez + c1.
g
y
Example 16. (i) In a two-dimensional fluid flow, the stream function is ψ = – ,
x + y2
2
.c
find the velocity potential φ. [M.T.U. 2014]

w
(ii) An electrostatic field in the xy-plane is given by the potential function φ = 3x2y – y3,
find the stream function and hence find complex potential. (G.B.T.U. 2011, 2013)
w
y
Sol. (i) ψ=– ...(1)
w
x + y2
2
∂ψ 2 xy ∂ψ y2 – x 2
= 2 , = 2
∂x (x + y2 )2 ∂y ( x + y 2 ) 2
∂φ ∂φ ∂ψ ∂ψ
We know that, dφ = dx + dy = dx – dy
∂x ∂y ∂y ∂x
( y2 − x 2 ) 2 xy
= dx − dy
(x2 + y2 )2 (x2 + y2 )2
( x 2 + y 2 ) dx − 2 x 2 dx − 2 xy dy
=
( x 2 + y2 )2
www.cgaspirants.com
www.cgaspirants.com
( x 2 + y 2 ) d( x) − x(2 x dx + 2 y dy)
=
(x 2 + y2 )2
( x 2 + y 2 ) d( x) − xd ( x 2 + y 2 ) x F I
=
(x2 + y2 )2
=d GH
x 2 + y2
. JK
x
Integration yields, φ= 2 + c where c is a constant.
x + y2
(ii) Let ψ (x, y) be the stream function.
∂ψ ∂ψ ∂φ FG
∂φ IJ FG IJ
dψ =
∂x
dx +
∂y
dy = −
∂y
dx +
H
∂x
dy
K H K
m
= {– (3x2 – 3y2)} dx + 6xy dy
= – 3x2 dx + (3y2 dx + 6xy dy)
o
= – d (x3) + 3d (xy2)
.c
Integrating, we get ψ = – x3 + 3xy2 + c |c is a constant
Complex potential is given by
or,
n
w = –i[x3 – iy3 + 3ix2y – 3xy2 – c]
ts
w = φ + iψ = 3x2y – y3 + i(–x3 + 3xy2 + c)
or, w = –i [(x + iy)3 – c]

ra
⇒ w = –iz3 + c1 | where c1 = ic
Example 17. (i) If u = ex(x cos y – y sin y) is a harmonic function, find an analytic
pi
function f (z) = u + iv such that f (1) = e.

(ii) Determine an analytic function f(z) in terms of z whose real part is e–x(x sin y – y cos y).
as
[M.T.U. 2012, G.B.T.U. 2011, U.P.T.U. 2006, 2008, 2014]

Sol. (i) We have, x
u = e (x cos y – y sin y)
g
∂u
= ex(x cos y – y sin y) + ex cos y = φ1 (x, y) |say
.c
∂x
∂u
w
= ex [– x sin y – y cos y – sin y] = φ2 (x, y) |say

∂y
∴ φ1(z, 0) = ez z + ez = (z + 1) ez
w
φ2 (z, 0) = 0
w
By Milne’s Thomson method,

f(z) = z {φ 1 ( z, 0) − i φ 2 ( z, 0)} dz + c | c is a constant
= z( z + 1) e z dz + c = (z – 1) ez + ez + c = zez + c
f(1) = e + c
...(1)
|From (1)
e=e+c |f(1) = e (given)
⇒ c=0
∴ From (1), f(z) = zez.
(ii) u = e–x(x sin y – y cos y)
∂u
= e–x sin y – e–x (x sin y – y cos y) = φ1 (x, y) | say
∂x
www.cgaspirants.com
www.cgaspirants.com
∂u
= e–x(x cos y – cos y + y sin y) = φ2(x, y) | say
∂y
∴ φ1 (z, 0) = 0 and φ2(z, 0) = e–z(z – 1)
f(z) = zl q
φ 1 ( z, 0) − iφ 2 ( z, 0) dz + c
=–i zL e − z ( z − 1) dz + c
NM
= − i ( z − 1) (− e − z ) −
= – i [(1 – z) e–z – e–z] + c

z OP
(− e − z ) dz + c
Q
m
⇒ f(z) = ize–z + c | where c is a constant
Example 18. (i) Determine the analytic function whose real part is e2x (x cos 2y – y sin 2y).
o
(ii) Find an analytic function whose imaginary part is e–x(x cos y + y sin y).
.c
(U.P.T.U. 2009)
Sol. (i) Let f(z) = u + iv be the required analytic function.
Here,
∴
∂u
∂x
u = e2x (x cos 2y – y sin 2y)
n ts
= e2x (2x cos 2y – 2y sin 2y + cos 2y) = φ1 (x, y) | say
ra
∂u
and = – e2x (2x sin 2y + sin 2y + 2y cos 2y) = φ2 (x, y) | say
∂y
Now, φ1 (z, 0) = e2z (2z + 1)
pi
φ2 (z, 0) = – e2z (0) = 0

as
f(z) = z {φ 1 ( z, 0) − i φ 2 ( z, 0)} dz + c = z e 2 z (2 z + 1) dz + c
z
g
e2 z e2 z
= (2z + 1) − 2. dz + c |Integrating by parts
.c
2 2
e 2 z 1 2z
w
= (2z + 1) − e +c
2 2
f(z) = ze2z + c where c is an arbitrary constant.
w
(ii) Let f(z) = u + iv be the required analytic function.

w
Here v = e–x(x cos y + y sin y)

∂v
= e–x (– x sin y + y cos y + sin y) = ψ1(x, y) | say
∂y
∂v
= e–x cos y – e–x (x cos y + y sin y) = ψ2 (x, y) | say
∂x
∴ ψ1 (z, 0) = 0
ψ2(z, 0) = e–z – e–z z = (1 – z) e–z
f(z) = z ψ 1 ( z, 0) + i ψ 2 ( z, 0) dz + c = i z (1 − z) e − z dz + c
www.cgaspirants.com
www.cgaspirants.com
LM
= i (1 − z) (− e − z ) −
N z (− 1) (− e − z ) dz + c OP
Q
= i [(z – 1) e–z + e–z] + c
⇒ f(z) = ize–z + c
Example 19. (i) Let f(z) = u(r, θ) + iv(r, θ) be an analytic function. If u = – r3 sin 3θ, then
construct the corresponding analytic function f(z) interms of z.
FG 1IJ
(ii) Find the analytic function f(z) = u + iv, given that v = r −
H r K
sin θ ; r ≠ 0
Sol. (i) u = – r3 sin 3θ
m
∂u ∂u
= – 3r2 sin 3θ, = – 3r3 cos 3θ
∂r ∂θ
o
∂v ∂v 1 ∂u FG ∂u IJ FG IJ
we know that dv = dr + dθ = − H
dr + r dθ K H K
.c
∂r ∂θ r ∂θ ∂r
= (3r2 cos 3θ) dr – (3r3 sin 3θ) dθ
⇒
Integration yields,
dv = d (r3 cos 3θ)
n ts
v = r3 cos 3θ + c
ra
∴ f(z) = u + iv = – r3 sin 3θ + ir3 cos 3θ + ic

pi
= ir3 (cos 3θ + i sin 3θ) + c1 | c1 = ic

= i(reiθ)3 + c1
as
⇒ f(z) = iz3 + c1 |∵ z = reiθ
FG 1IJ sin θ
H rK
v= r−
g
(ii)
.c
∂v F 1I
= G 1 + J sin θ,
∂v 1 FG IJ
H rK = r−
H
cos θ
K
w
∂r 2 ∂θ r
we know that,
w
∂u ∂u 1 ∂v FG ∂v IJ FG IJ
du = dr + dθ =
r ∂θ H
dr + − r
∂r
dθK H K
w
∂r ∂θ
FG IJ cos θ dr – FG r + 1IJ sin θ dθ

1
= 1−
H r K 2 H rK
F1 I
du = d (r cos θ) + d G cos θJ
⇒
Hr K
F 1I
u = G r + J cos θ + c
Integration yields,
H rK
F 1I F 1I
f(z) = u + iv = G r + J cos θ + c + i G r − J sin θ
∴
H rK H rK
www.cgaspirants.com
www.cgaspirants.com
1 –iθ
= reiθ + e +c
r
1
⇒ f (z) = z +
+ c.
z
Example 20. If u – v = (x – y) (x2 + 4xy + y2) and f(z) = u + iv is an analytic function of
z = x + iy, find f(z) in terms of z.
Sol. Here, f(z) = u + iv
∴ if(z) = iu – v
Adding (1 + i) f(z) = (u – v) + i(u + v)
Let (1 + i) f(z) = F(z), u – v = U, u + v = V, then
m
F(z) = U + iV
o
Now, U = u – v = (x – y) (x2 + 4xy + y2)
∂U
.c
⇒ = x2 + 4xy + y2 + (x – y)(2x + 4y) = 3x2 + 6xy – 3y2 = φ1(x, y)
∂x
| say
and
∂U
∂y
n ts
= – (x2 + 4xy + y2) + (x – y)(4x + 2y) = 3x2 – 6xy – 3y2 = φ2(x, y)
| say
ra
Now, φ1(z, 0) = 3z2, φ2(z, 0) = 3z2

z
pi
F(z) = z [φ 1 ( z, 0) – iφ 2 ( z, 0)] dz + c = [3 z 2 − i(3 z 2 )] dz + c

as
F(z) = (1 – i) z3 + c
⇒ (1 + i) f(z) = (1 – i) z3 + c
g
FG 1 − i IJ z c FG IJ
− 2i 3 FG where c c IJ
.c
3
or, f(z) =
H 1 + iK +
1+ i
=
H K
2
z + c1
H 1 =
1+ i K
w
or, f(z) = – iz3 + c1.

2 sin 2x
w
Example 21. If u + v = 2y
and f(z) = u + iv is an analytic function of
e + e −2y − 2 cos 2x
w
z = x + iy, find f(z) in terms of z.

Sol. Let f(z) = u + iv ...(1)
Multiplying both sides by i
i f(z) = iu – v ...(2)
Adding (1) and (2), we get
(1 + i) f(z) = (u – v) + i(u + v) ...(3)
⇒ F(z) = U + iV ...(4)
where F(z) = (1 + i) f(z) ...(5)
U=u–v and V=u+v ...(6)
www.cgaspirants.com
www.cgaspirants.com
It means that we have been given
sin 2 x
V= ...(7) | ∵ e2y + e–2y = 2 cosh 2y
cosh 2 y − cos 2 x
∂V − 2 sin 2 x sinh 2 y
Now = = ψ1(x, y) | say
∂y (cosh 2 y − cos 2 x) 2
∂V 2 cos 2 x (cosh 2 y − cos 2 x) − 2 sin 2 2 x

and =
∂x (cosh 2 y − cos 2 x) 2
2 cos 2 x cosh 2 y − 2
m
= = ψ2(x, y) | say
(cosh 2 y − cos 2 x) 2
o
∴ ψ1(z, 0) = 0
.c
2(cos 2 z − 1) −2 −2
ψ2(z, 0) = 2
= = = – cosec2 z
(1 − cos 2 z) 1 − cos 2 z 1 − 1 + 2 sin 2 z
By Milne’s Thomson method, we have
n ts
F(z) = ∫ {ψ1(z, 0) + i ψ2(z, 0)} dz + c
ra
= ∫ – i cosec2 z dz + c = i cot z + c
Replacing F(z) by (1 + i) f(z), from eqn. (5), we get
pi
(1 + i) f(z) = i cot z + c
i c
as
⇒ f(z) = cot z +
1+ i 1+ i
1
c
∴ f(z) = (1 + i) cot z + c1, where c1 = .
g
2 1+ i
.c
cos x + sin x − e − y
Example 22. If f(z) = u + iv is an analytic function of z and u – v = ,
2 cos x − 2 cosh y
w
1 LM z OP π FG IJ
prove that f(z) =
2
1 − cot
N 2 Q
when f
2 H K
= 0.
w
Sol. Let f(z) = u + iv ...(1)

w
∴ i f(z) = iu – v
Add, (1 + i) f(z) = (u – v) + i(u + v) ...(2)
⇒ F(z) = U + iV ...(3)
where u – v = U, u + v = V and (1 + i) f(z) = F(z).
cos x + sin x − e − y
We have, u–v=
cos x + sin x − cosh y + sinh y
or, U= [∵ e–y = cosh y – sinh y]
1 sin x + sinh y
= + ...(4)
2 2(cos x − cosh y)
www.cgaspirants.com
www.cgaspirants.com
Diff. (4) w.r.t. x partially, we get

LM
∂U 1 (cos x − cosh y) cos x − (sin x + sinh y)(− sin x)
=
OP
∂x 2 N (cos x − cosh y) 2 Q
φ1(x, y) =
LM OP
1 1 − cosh y cos x + sinh y sin x
N2 (cos x − cosh y) 2 Q
1 L 1 − cos z O
φ (z, 0) = M
1 P= 1 .
2 N (cos z − 1) Q 2(1 − cos z) 2 ...(5)
Diff. (4) partially w.r.t. y, we get
m
LM
∂U 1 (cos x − cosh y) . cosh y − (sin x + sinh y)(− sinh y)
=
OP
∂y 2 N (cos x − cosh y) 2 Q
o
1 L cos x cosh y + sin x sinh y − 1 O
.c
φ (x, y) = M PQ
2
2N (cos x − cosh y) 2
∴ 2
1 L cos z − 1 O
φ (z, 0) = M P
2 N (cos z − 1) Q
1 F −1 I
= .G J.
2 H 1 − cos z K
n
2
ts ...(6)
ra
By Milne’s Thomson Method,

F(z) = zL [φ 1 ( z, 0) − i φ 2 ( z, 0)] dz + c
pi
= z MN 1
.
1 i
+ .
1
dz + c
OP
Q
as
2 (1 − cos z) 2 1 − cos z
=
1+ i 1
z
dz + c =
1+ i
cosec 2 ( z/2) dz + c z
g
2
2 2 sin z/2 4
FG IJ FG IJ
.c
1 + i (− cot z/2) 1+ i z
=
H
4
.
1 K
+c =–
FG IJ
2
cot + c
2 H K
H K
w
2
1+ i FG z IJ
w
or, (1 + i) f(z) = –
2 H
cot + c
2 K
1 z c
w
⇒ f(z) = – cot + ...(7)

2 2 1+ i
FG π IJ = − 1 cot π + c
f
H 2K 2 4 1 + i [From (7)]
1 c c 1
0=– + ⇒ = ...(8)
2 1+ i 1+ i 2
1 z 1 1 FG
z IJ
∴ From (7), f(z) = –
2
cot + =
2 2 2
1 − cot
2H.
K [Using (8)]
www.cgaspirants.com
www.cgaspirants.com
Example 23. (i) If f(z) is a regular function of z, prove that

F∂ 2
∂2 I |f(z)|
GH ∂x 2
+
∂y 2 JK 2 = 4 | f ′(z)|2. (U.P.T.U. 2007, 2015)
(ii) If f(z) is a harmonic function of z, show that
RS ∂ |f(z)|UV + RS ∂ |f(z)|UV
2 2
= f ′ (z) 2
(U.P.T.U. 2009)
T ∂x W T ∂y W
Sol. (i) Let f(z) = u + iv so that |f(z)| = u 2 + v2
or |f(z)|2 = u2 + v2 = φ(x, y) (say)
m
∂φ ∂u ∂v
∴ = 2u + 2v
∂x ∂x ∂x
o
∂2φ LM
∂ 2u ∂u FG IJ 2
FG IJ
∂ 2v ∂v
2 OP
.c
∂x 2
=2 u 2 +
MN
∂x ∂x H K +v
H K
∂x 2
+
∂x PQ
L ∂ u F ∂u I OP
Similarly,
∂ φ
∂y
2
= 2 Mu
MN ∂y + GH ∂y JK
2
n 2
2
2
+v
∂y
+G J ts
∂ v F ∂v I
2
H ∂y K 2
2
PQ
Adding, we get
ra
∂2φ ∂2φ LM F
∂ 2u ∂ 2u ∂u I FG IJ + F ∂u I 2 2
F ∂ v + ∂ v I + FG ∂v IJ + F ∂v I OP
2 2 2 2
+ = 2 u
MN GH+ 2 + JK H K GH ∂y JK H ∂x ∂y JK H ∂x K GH ∂y JK PQ
+ vG ...(1)
pi
2 2 2 2 2
∂x ∂y ∂x ∂y ∂x
Since f(z) = u + iv is a regular function of z, u and v satisfy C-R equations and Laplace’s
as
equation.
∂u ∂v ∂u ∂v ∂ 2u ∂ 2u ∂ 2v ∂ 2v
= , =− and + = 0 = +
g
∴
∂x ∂y ∂y ∂x ∂x 2 ∂y 2 ∂x 2 ∂y 2
.c
∴ From (1), we get

LM F IJ + FG ∂vIJ FG ∂vIJ + FG ∂u IJ OP
w
2 2 2 2
∂2φ ∂2φ ∂u
MN GH K H ∂x K H ∂x K H ∂x K PQ
2
+ 2 =2 0+ +0+
∂x ∂y ∂x
w
LF ∂u I 2
F ∂vI O 2
= 4 MG J +G J P
w
MNH ∂x K H ∂x K PQ ...(2)
Now f(z) = u + iv
∂u ∂v
|f ′(z)|2 =
FG ∂u IJ + FG ∂v IJ
2 2
∴ f ′(z) =
∂x
+i
∂x
and
H ∂x K H ∂x K
From (2), we get
F∂ 2
∂2 I φ = 4 |f ′(z)| F∂ 2
∂2 I |f(z) |
GH ∂x 2
+
∂y 2
JK 2 or GH ∂x 2
+
∂y 2
JK 2 = 4 |f ′(z)|2.
www.cgaspirants.com
www.cgaspirants.com
(ii) We have, f(z) = u + iv ...(1)
∴ |f(z)| = u 2 + v2 ...(2)
Partially differentiating eqn. (2) w.r.t. x and y, we get
∂u ∂v
∂ 1 2 2 −1/2 ∂u ∂vFG IJ u
∂x
+v
∂x
∂x
|f ( z)| = (u + v )
2
2u
∂x
+ 2v
∂x H
=
K |f ( z)|
...(3)
∂u ∂v
u +v
∂ ∂y ∂y
Similarly, |f ( z)| = ...(4)
∂y |f ( z)|
m
Squaring and adding (3) and (4), we get
FG u ∂u + v ∂v IJ + FG u ∂u + v ∂vIJ
2 2
o
RS |f (z)|UV + S |f (z)|V = H ∂x ∂x K H ∂y ∂y K
∂
2
R ∂ U 2
.c
T ∂x W T ∂y W |f ( z)| 2
FG u ∂u + v ∂vIJ + FG − u ∂v + v ∂u IJ
=
n 2
|f ( z)|2
ts 2
| Using C-R eqns.

ra
(u 2 + v2 )
LMFG ∂u IJ + FG ∂v IJ OP
2 2
MNH ∂x K H ∂x K PQ
pi
=
|f ( z)|2
as
FG ∂u IJ + FG ∂v IJ
2 2
=
H ∂x K H ∂x K |∵ |f(z)|2 = u2 + v2
g
∂u ∂v
= |f ′(z)|2 ∵ f ′ ( z) = +i
.c
∂x ∂x
w
ASSIGNMENT
w
1. (i) Determine a, b, c, d so that the function f(z) = (x2 + axy + by2) + i(cx2 + dxy + y2) is analytic.
(ii) Find the constants a, b, c such that the function f(z) where
w
f(z) = – x2 + xy + y2 + i (ax2 + bxy + cy2) is analytic. Express f(z) in terms of z.

(M.T.U. 2013)
(iii) Find the value of the constants a and b such that the following function f(z) is analytic.
f(z) = cos x (cosh y + a sinh y) + i sin x (cosh y + b sinh y)
1 px
(iv) Determine p such that the function f(z) = log (x2 + y2) + i tan–1 is an analytic function.
2 y
Also find f ′(z). (M.T.U. 2012)
2. Show that
(a) f(z) = xy + iy is everywhere continuous but is not analytic.
(b) f(z) = z + 2 z is not analytic anywhere in the complex plane.
(c) f(z) = z | z | is not analytic anywhere. (U.K.T.U. 2010)
www.cgaspirants.com
www.cgaspirants.com
3. Discuss the analyticity of the following functions:

1
(i) sin z (ii) cosh z (iii) (iv) z3.
z
4. (i) Define analytic function. Discuss the analyticity and differentiability of f(z) = | z |4 at z = 0.
(G.B.T.U. 2012)
(ii) Define anlaytic function. Discuss the analyticity of f(z) = Re (z3) in the complex plane.
(U.P.T.U. 2014)
∂u 1 ∂v ∂v 1 ∂u
5. Show that the polar form of Cauchy-Riemann equations are = , =− . Deduce
∂ 2u 1 ∂u 1 ∂ 2u
that 2
++ 2 = 0.
∂r r ∂r r ∂θ2
m
6. (i) Show that an analytic function f(z), whose derivative is identically zero, is constant.
(ii) It is given that a function f(z) and its conjugate f ( z) are both analytic. Determine the function
o
f(z).
(iii) Show that if f(z) is analytic and Im f (z) = constant then f (z) is a constant.
.c
ux uy
(iv) Show that if f(z) is differentiable at a point z, then |f ′(z)|2 = v vy
ts
x
x3 y5 ( x + iy)
7. (i) Show that the function f(z) defined by f(z) = , z ≠ 0, f(0) = 0, is not analytic at the
n
x6 + y10
origin even though it satisfies Cauchy-Riemann equations at the origin. (G.B.T.U. 2011)
ra
(ii) Show that for the function

R| (z) ,
2
pi
f(z) = S| z z≠0
T 0, z=0
as
the Cauchy-Riemann equations are satisfied at the origin. Does f ′(0) exist?
(iii) Show that for the function
R| 2 xy (x + iy) ,
g
z≠0
f(z) = S| x + y
2 2
.c
T 0, z=0
w
the C-R equations are satisfied at origin but derivative of f(z) does not exist at origin.
8. (i) If u is a harmonic function then show that w = u2 is not a harmonic function unless u is a
w
constant.
(ii) If f(z) is an analytic function, show that |f (z)| is not a harmonic function.
w
9. (i) Show that the function u (x, y) = 2x + y3 – 3x2y is harmonic. Find its conjugate harmonic
function v(x, y) and the corresponding analytic function f(z).
(ii) Show that the function v(x, y) = ex sin y is harmonic. Find its conjugate harmonic function
u(x, y) and the corresponding analytic function f(z).
(iii) Define a harmonic function and conjugate harmonic function. Find the harmonic conjugate of
the function u(x, y) = 2x (1 – y). (U.P.T.U. 2009)
(iv) Show that the function u = e –2xy 2 2
sin (x – y ) is harmonic. (U.K.T.U. 2011)
(v) Show that u(x, y) = x3 – 4xy – 3xy2 is harmonic. Find its harmonic conjugate v(x, y) and the
corresponding analytic function f(z) = u + iv. (G.B.T.U. 2013)
10. (i) Show that the function u(r, θ) = r2 cos 2θ is harmonic. Find its conjugate harmonic function
and the corresponding analytic function f(z).
(ii) Determine constant ‘b’ such that u = ebx cos 5y is harmonic.
www.cgaspirants.com
www.cgaspirants.com
(iii) Define Harmonic function. Show that the function v = log (x2 + y2) + x – 2y is harmonic. Also
find the analytic function f(z) = u + iv. (G.B.T.U. 2012)
(iv) Show that v(x, y) = e–x (x cos y + y sin y) is harmonic. Find its harmonic conjugate.
(U.P.T.U. 2014)
11. Determine the analytic function f(z) in terms of z whose real part is
1
(i) log (x2 + y2) (U.K.T.U. 2011) (ii) cos x cosh y
2
(iii) e–x (x cos y + y sin y) ; f (0) = 1 (iv) (x – y)(x2 + 4xy + y2) (G.B.T.U. 2012)
sin 2 x sin 2 x
(v) (vi) .
cosh 2 y − cos 2 x cosh 2 y + cos 2 x
12. Find the regular function f(z) in terms of z whose imaginary part is
x− y
m
(i) 2 (ii) cos x cosh y (iii) sinh x cos y
x + y2
x
o
(iv) 6xy – 5x + 3 (v) 2 + cosh x cos y. (vi) ex (x sin y + y cos y)
x + y2
(U.P.T.U. 2015)
.c
13. Prove that u = x2 – y2 – 2xy – 2x + 3y is harmonic. Find a function v such that f(z) = u + iv is
analytic. Also express f(z) in terms of z.
14.
stream function.
n ts
(i) An electrostatic field in the xy-plane is given by the potential function φ = x2 – y2, find the
(ii) If the potential function is log (x2 + y2), find the flux function and the complex potential
ra
function.
15. (i) In a two dimensional fluid flow, the stream function is ψ = tan–1
FG y IJ , find the velocity
H xK
pi
potential φ.
x
(ii) If w = φ + iψ represents the complex potential for an electric field and ψ = x2 – y 2 + 2 ,
as
determine the function φ.

x + y2
(iii) If u = (x – 1)3 – 3xy2 + 3y2, determine v so that u + iv is a regular function of x + iy.
g
[U.K.T.U. 2010]
F∂ 2
∂2 I |Re f(z)|
.c
16. If f(z) is an analytic function of z, prove that GH ∂x 2

+
∂y 2 JK 2 = 2 |f ′(z)|2.
w
(G.B.T.U. 2012)
17. Find an analytic function f(z) = u(r, θ) + iv(r, θ) such that v(r, θ) = r2 cos 2θ – r cos θ + 2.
w
18. If f(z) = u + iv is an analytic function, find f(z) in terms of z if

x
w
(i) u – v = ex (cos y – sin y) (ii) u + v = , when f(1) = 1

x 2 + y2
[U.P.T.U. (C.O.) 2008]
y
e − cos x + sin x π FG IJ
3−i
(iii) u – v =
cosh y − cos x
when f
2
=
H K
2
.
19. (i) If f(z) = u + iv is an analytic function of z = x + iy and u + v = (x + y) (2 – 4xy + x2 + y2), then

construct f(z) in terms of z.
(ii) If f(z) = u + iv is an analytic function of z = x + iy and u – v = e –x [(x – y) sin y – (x + y) cos y], then
construct f(z) in terms of z. [U.P.T.U. (C.O.) 2009]
20. If f = u + iv is analytic show that g = – v + iu is also analytic. Also show that u and – v are
conjugate harmonic.
www.cgaspirants.com
www.cgaspirants.com
21. Show that the function

z
(i) f (z) = is analytic at z = ∞. (ii) f(z) = z is not analytic at z = ∞.
z+1
z+z z−z
22. If f (z) = u(x, y) + iv (x, y) where x = ,y= is continuous as a function of two variables z
2 2i
∂f
and z then show that = 0 is equivalent to the Cauchy-Riemann equations.
∂z
LMHint. ∂f = F ∂u ∂x + ∂u ∂y I + i F ∂v ∂x + ∂v ∂y I OP
MN ∂z GH ∂x ∂z ∂y ∂z JK GH ∂x ∂z ∂y ∂z JK PQ
m
∂ 2u
23. (i) Show that a harmonic function satisfies the formal differential equation =0
∂z ∂ z
o
F∂ 2
∂2 I log |f ′(z)| = 0. Further, if |f ′(z)|
(ii) If w = f(z) is a regular function of z, prove that GH ∂x + JK
.c
2 2
∂y
is the product of a function of x and function of y, show that f ′(z) = exp. (αz2 + βz + γ) where α
is real and β, γ are complex constants.
(i) 3u + v = 3 sin x cosh y + cos x sinh y

n ts
24. If f (z) = u + iv is an analytic function of z = x + iy, find f (z) in terms of z if
(ii) u – 2v = cos x cosh y + 2 sin x sinh y
(iii) 2u – v = ex (2 cos y – sin y)
ra
25. (i) If f ′(z) = f(z) for all z, then show that f(z) = ke z, where k is an arbitrary constant.
(ii) Find an analytic function f(z) such that Re [f ′(z)] = 3x2 – 4y – 3y2 and f(1 + i) = 0
pi
(iii) Let f(z) = u + iv and g(z) = v + iu be analytic functions for all z. Let f(0) = 1 and g(0) = i. Obtain
the value of h(z) at z = 1 + i where h(z) = f ′(z) + g′(z) + 2f(z) g(z).
as
(iv) If f(z) = u + iv is an analytic function of z and φ is a function of u and v, then show that
FG ∂φ IJ + FG ∂φ IJ = LMFG ∂φ IJ + FG ∂φ IJ OP|f ′ (z)|

2 2 2 2
2
H ∂x K H ∂y K MNH ∂u K H ∂v K PQ
g
.c
Answers
1 1 1
w
1. (i) a = 2, b = – 1, c = – 1, d = 2 (ii) a = − , b = – 2, c = ; f(z) = − (2 + i) z2

2 2 2
1
w
(iii) a = – 1, b = – 1 (iv) p = – 1, f ′(z) =

z
6. (ii) constant function 7. (ii) No
w
9. (i) v = 2y – 3xy2 + x3 + c ; f(z) = 2z + iz3 + ic (ii) u = ex cos y + c ; f(z) = ez + c

(iii) v(x, y) = x2 – y2 + 2y + c (v) v(x, y) = 2x2 – 2y2 + 3x2y – y3 + c, f(z) = z3 + 2iz2 + c
10. (i) v = r2 sin 2θ + c ; f(z) = z2 + ic (ii) b = ± 5
(iii) f(z) = – 2z + i(2 log z + z) + c (iv) u(x, y) = e–x (x sin y – y cos y) + c
11. (i) log z + c (ii) cos z + c (iii) 1 + ze–z
(iv) (1 – i) z3 + c (v) cot z + c (vi) tan z + c
1+ i
12. (i) +c (ii) i cos z + c (iii) i sinh z + c
z
i
(iv) 3z2 – 5iz + c (v) + i cosh z + c (vi) z ez + c
z
www.cgaspirants.com
www.cgaspirants.com
13. v = x2 – y2 + 2xy – 2y – 3x + c, f(z) = (1 + i)z2 – (2 + 3i)z + ic

FG y IJ + c, 2 log z + ic
14. (i) ψ = 2xy + c (ii) 2 tan–1
H xK
1 y
15. (i) log (x2 + y 2) + c (ii) – 2xy + +c (iii) v = 3y (1 + x2) – y3
2 x + y2
2
17. i(z2 – z + 2) + c
1 FG i + 1IJ z 1
18. (i) ez + c (ii)
1+ i Hz K (iii) cot + (1 – i)
2 2
21. (i) 2z + iz3 + c (ii) ize–z + c
24. (i) f (z) = sin z + c (ii) f (z) = cos z + c (iii) f (z) = ez + c
m
25. (ii) f(z) = z3 + 2iz2 + 6 – 2i (iii) 2i
1.19 LINE INTEGRAL IN THE COMPLEX PLANE
o
z
.c
b
In case of real variable, the path of integration of f ( x) dx is always along the x-axis from
a
x = a to x = b. But in case of a complex function f(z), the path of the definite integral
can be along any curve from z = a to z = b.
n ts za
b
f ( z) dz
ra
Let f(z) be a continuous function of the complex variable z = x + iy defined at all points
of a curve C having end points A and B. Divide the curve C into n parts at the points
A = P0(z0), P1(z1), ......, Pi(zi), ......, Pn(zn) = B.
pi
Let δzi = zi – zi–1 and ξi be any point on the arc Pi–1 Pi. Then the limit of the sum
n
∑ f (ξ ) δz
as
i i as n → ∞ and each δzi → 0, if it exists, is called the line integral of f(z)
z
i=1
g
along the curve C. It is denoted by f ( z) dz .

C
.c
In case the points P0 and Pn coincide so that

Y Pn = B
C is a closed curve, then this integral is called
z
w
Pi +1
contour integral and is denoted by f ( z) dz .
C
w
If f(z) = u(x, y) + iv(x, y), then since dz = dx + i dy, Pi

we have
z z
w
Pi – 1
f ( z) dz = (u + iv)(dx + i dy) P2
C C P1
= z
C
(udx − vdy) + i z
C
which shows that the evaluation of the line integral
(vdx + udy)
P0 = A
O X
of a complex function can be reduced to the evalua-
tion of two line integrals of real functions.
Moreover, the value of the integral depends on the path of integration unless the
integrand is analytic.
www.cgaspirants.com
www.cgaspirants.com
When the same path of integration is used in each integral, then
za
b
f ( z) dz = −
If c is a point on the arc joining a and b, then

z
b
a
f ( z) dz
za
b
f ( z) dz = z c
a
f ( z) dz + z
c
b
f ( z) dz .
EXAMPLES
z 1+ i
m
Example 1. Evaluate (x − y + ix 2 ) dz .
0
(a) along the straight line from z = 0 to z = 1 + i
o
(b) along the real axis from z = 0 to z = 1 and then along a line parallel to imaginary axis
.c
from z = 1 to z = 1 + i.
(c) along the imaginary axis from z = 0 to z = i and then along a line parallel to real axis
from z = i to z = 1 + i.
Sol. (a) Along the straight line OP joining O(z = 0) and P(z = 1 + i), y = x, dy = dx and x
varies from 0 to 1.
n ts
z z
ra
1+ i 1+ i
∴ ( x − y + ix 2 ) dz = (x – y + ix2)(dx + i dy)
0 0
z z
pi
1 1
= ( x − x + ix 2 )(dx + idx) = ix 2 (1 + i) dx
0 0
as
Fx I 3 1
1 1
= (i – 1) G J =− + i.
H 3K 0
3 3
g
(b) Along the path OAP where A is z = 1

.c
z 1+ i
( x − y + ix 2 ) dz
w
z ( x − y + ix 2 ) dz + z
w
= (x – y + ix2) dz ...(1)
OA AP
Now along OA, y = 0, dz = dx and x varies from 0 to 1.

w
z z LM x OP
2 1
1 x3 1 1
∴ ( x − y + ix 2 ) dz = ( x + ix 2 ) dx = +i = + i
OA 0
N2 3 Q 0
2 3
Also along AP, x = 1, dz = idy and y varies from 0 to 1
∴ zAP
2 L
N z0
1
( x − y + ix ) dz = (1 − y + i) idy = M(− 1 + i) y − i
y O
2
P
Q
1 1
=–1+i– i=–1+ i
2 2
2 1
Hence from (1),

z 0
F1 1 I F 1 I 1 5
1+ i
( x − y + ix ) dz = G + iJ + G − 1 + iJ = − + i .
H2 3 K H 2 K 2 6
2
www.cgaspirants.com
www.cgaspirants.com
(c) Along the path OBP where B is z = i
z 0
1+ i
( x − y + ix 2 ) dz =
Now along OB, x = 0, dz = idy and y varies from 0 to 1

z OB
( x − y + ix 2 ) dz + zBP
(x – y + ix2) dz ...(2)
∴ zOB
( x − y + ix 2 ) dz = z 1
0
(− y) idy = − i
LM y OP
2 1
N2Q 0
=−
1
2
i
Also, along BP, y = 1, dz = dx and x varies from 0 to 1
∴ z ( x − y + ix 2 ) dz = z LM x − x + i x OP
1
N2
( x − 1 + ix 2 ) dx =
3Q
2 3 1
=−
1 1
+ i
2 3
m
BP 0
0
Hence from (2), z 1+ i

2
1 F 1 1I 1 1
(x – y + ix ) dz = – i + G − + iJ = − − i .
H 2 3K 2 6
o
0 2
.c
Note. The values of the integral are different along the three different paths.
Example 2. Evaluate
(a) y = x
z
(b) y = x2.
0
1+ i
n (x 2 − iy) dz along the paths
ts (G.B.T.U. 2010)
Sol. (a) Along the line y = x,
ra
dy = dx so that dz = dx + idx = (1 + i) dx
z 1+ i
( x 2 − iy) dz = z 1
( x 2 − ix)(1 + i) dx
pi
0 0
Lx
= (1 + i) M
3
x2 OP 1
FG 1 − 1 iIJ = 5 − 1 i .
as
N3
−i
2 Q 0
= (1 + i)
H3 2 K 6 6
(b) Along the parabola y = x2, dy = 2x dx so that
g
Y
dz = dx + 2ix dx = (1 + 2ix) dx
.c
and x varies from 0 to 1.
z z
w
1+ i 1 P (1, 1)
∴ ( x 2 − iy) dz = ( x 2 − ix 2 ) (1 + 2ix) dx
0 0
w
L x + i x OP 4 1
x
3
=
= (1 − i) M
w
N3 2Q
2
x
0
=
y
F1 1 I 5 1
= (1 – i) GH + iJK = + i . O
3 2 6 6 X
Example 3. Evaluate z0
2+i
(z) 2 dz , along
(a) the real axis from z = 0 to z = 2 and then along a line parallel to y-axis from z = 2 to
z = 2 + i. (U.P.T.U. 2009, U.K.T.U. 2011)
(b) along the line 2y = x. (U.P.T.U. 2009)
www.cgaspirants.com
www.cgaspirants.com
Sol. ( z )2 = (x – iy)2 = (x2 – y2) – 2ixy Y

(a) Along the path OAP where A is (2, 0) and P is
(2, 1).
z
P (2, 1)
2+ i
2
( z ) dz
z z
0 x
=
= ( x 2 − y 2 − 2ixy) dz + ( x 2 − y 2 − 2ixy) dz 2y
OA AP
...(1)
Now, along OA, y = 0, dz = dx and x varies from 0
to 2 O A (2, 0) X
z z 2 LM x OP 3 2
8
m
∴ ( x 2 − y 2 − 2ixy) dz = x 2 dx = =
OA 0
N3Q 0
3
Also, along AP, x = 2, dz = idy and y varies from 0 to 1
o
z zL 1
.c
∴ ( x 2 − y 2 − 2ixy) dz = (4 – y2 – 4iy) idy
AP 0
y3 OP 1
MN
= 4iy − i
3
+ 2 y2
Q ts
= 4i −
1
3
i+2=2+
11
3
i
z
0
n
8 11 14 11 2+i
2
Hence from (1), we have ( z ) dz = + 2 + i= + i.
ra
0 3 3 3 3
(b) Along the line OP, 2y = x, dx = 2dy
so that dz = 2dy + i dy = (2 + i) dy
pi
and y varies from 0 to 1.
z 2+ i
( z ) 2 dz = z 2+ i
( x 2 − y 2 − 2ixy) dz = z 1
(4 y 2 − y 2 − 4iy 2 ) (2 + i) dy
as
∴
0 0 0
= (2 + i)(3 – 4i) z 1
y 2 dy = (10 − 5i)
LM y OP
3 1
10 5
− i.
g
=
0
N3Q 0
3 3
.c
Example 4. Integrate f(z) = x2 + ixy from A(1, 1) to B(2, 4) along the curve x = t, y = t2.
Sol. Equations of the path of integration are x = t, y = t2
w
∴ dx = dt, dy = 2t dt
At A(1, 1), t = 1 and at B(2, 4), t = 2
w
∴ z B
f ( z) dz = z B
( x 2 + ixy)(dx + idy) = z 2
(t2 + it3)(dt + 2it dt)
w
A A 1
=
z1
2
(t − 2t ) dt + i 3t dt = M −
2 4
z
L t 2t OP + i LM 3t OP
N3 5 Q N 4 Q
1
2
3
3 5 2
1
4 2
1
F 8 64 IJ − FG 1 − 2 IJ + i FG 12 − 3 IJ = − 151 + 45 i .
=G −
H 3 5 K H 3 5K H 4 K 5 4
Example 5. Prove that
(i)
z dz
z−a
= 2πi
z
C
(ii) (z − a) n dz = 0 [n is an integer ≠ –1] where C is the circle | z – a | = r.

C
(G.B.T.U. 2011)
www.cgaspirants.com
www.cgaspirants.com
Sol. The equation of the circle C is

|z–a|=r or z – a = reiθθ
where θ varies from 0 to 2π as z describes C once in the anti-clockwise direction.
Also dz = ireiθ dθ.
(i)
zC
dz
z−a
= z0
2π ire iθ dθ
re iθ
=i z 2π
0
dθ = 2πi
(ii)
z
C
( z − a) n dz = z0
2π
r n e niθ . ire iθ dθ
= ir
n+1
z 2π
e i( n + 1)θ dθ
m
0
LM e OP
i( n + 1)θ 2π
o
= irn+1 [∵ n ≠ – 1]
N i(n + 1) Q 0
.c
r n+ 1 i2(n+1)π
= [e – 1]
= 0.
n+1
[∵ ts
ei2(n+1)π = cos 2(n + 1)π + i sin 2(n + 1)π = 1]
z
n
Example 6. Evaluate the integral |z|dz , where c is the contour
ra
c
(i) The straight line from z = – i to z = i
(ii) The left half of the unit circle | z | = 1 from z = – i to z = i.
pi
Sol. (i) The straight line from z = – i to z = i is x = 0

i.e., z = iy so that dz = idy
z z z z
as
1 0 1
∴ |z|dz = |iy|i dy = i (− y) dy + i y dy
c −1 −1 0
g
Fy I
=–iG J
2 0
Fy I
+iG J
2 1
FG 1IJ + i FG 1IJ = i.
=–i −
H 2K H 2K
.c
H 2K −1
H 2K 0
(ii) For a point on the unit circle | z | = 1,
w
z = eiθ
w
∴ dz = ieiθ dθ.
3π π
The points z = – i and i correspond to θ = and θ = respectively.
w
2 2
z z F I π/2
idθ = GH e JK
π/2 iθ
iθ
∴ | z| dz = 1. e = eiπ/2 – e3iπ/2
c 3π / 2 3π / 2
π π 3π 3π
= cos + i sin − cos − i sin = 0 + i – 0 – i(– 1) = 2i.
2 2 2 2
Example 7. Evaluate the integral
Sol. Here, c ≡ |z| = 1
z c
log z dz , where c is the unit circle | z | = 1.
...(1)
z c
log zdz = z c
log ( x + iy) dz
www.cgaspirants.com
www.cgaspirants.com
= z LMN
c
1
2
log ( x 2 + y 2 ) + i tan −1
y
x
dz
OP
Q
On the unit circle,

=i
z = eiθ
z c
tan −1
FG y IJ dz.
H xK ...(2) (∵ x2 + y2 = 1)
∴ dz = ieiθ dθ.
Now (2) becomes,
z c
log z dz = i z 2π
0
tan–1 (tan θ) ieiθ dθ = –
z
0
2π
θ e iθ d θ
LF e I OP LM F I OP
z
2π 2π
m
= – MG θ
iθ 2π e iθ 2π 2 πi 1 e iθ
MNH i JK − 0
0
1.
i
dθ = –
PQ i
e −
MN
i i GH JK P
0Q
o
=– M
L 2π e + e OP
.c
2 πi 2 πi
− 1 = 2πie2πi + 1 – e2πi = 2πi |∵ e2πi = 1
Ni Q
n ASSIGNMENT ts
z
ra
3+ i
1. Evaluate z2 dz , along
0
pi
x
(a) the line y = (b) the real axis to 3 and then vertically to 3 + i
3
as
(c) the parabola x = 3y2.
2. Find the value of the integral

z 1+ i
( x − y − ix2 ) dz , along real axis from z = 0 to z = 1 and then
g
0
along a line parallel to imaginary axis from z = 1 to z = 1 + i. [G.B.T.U. (C.O.) 2011]
.c
3. Evaluate
z 4 + 2i
z dz along the curve given by z = t2 + it.
w
4. (i) Evaluate
z | z |2 dz around the square with vertices at (0, 0), (1, 0), (1, 1) and (0, 1).
w
z
w
(ii) Show that ( z + 1) dz = 0 where C is the boundary of the square whose vertices are at the
C
points z = 0, z = 1, z = 1 + i and z = i.
5. (a) Evaluate
z C
( x + y) dx + x 2 y dy
(i) along y = x2 having (0, 0), (3, 9) as end points.

(ii) along y = 3x between the same points.
(b) Evaluate
(i) along y = x2
z (1, 1)
(0, 0)
(3 x2 + 4 xy + 3 y2 ) dx + 2(x2 + 3xy + 4y2)dy
(ii) along y = x
Does the value of the integral depend upon the path?
www.cgaspirants.com
www.cgaspirants.com
6. (i) Evaluate
z C
( y − x − 3 x 2 i) dz where C is the straight line from z = 0 to z = 1 + i.
(ii) Evaluate
z 2 + 3i
1− i
( z2 + z) dz along the line joining the points (1, – 1) and (2, 3).
7. (i) Evaluate
z C
( z − z2 ) dz where C is the upper half of the circle | z | = 1. What is the value of
this integral if C is lower half of the given circle?
(ii) Evaluate the integral

zC
( z − z2 ) dz where C is the upper half of the circle | z – 2 | = 3. What is
the value of the integral if C is the lower half of the circle? (M.T.U. 2013)
m
[Hint: z = 2 + 3eiθ]
8. Prove that
1
z
dz = − πi or πi according as C is the semi-circular arc | z | = 1 from z = – 1
o
C z
to z = 1 above or below the real axis.
.c
9. Evaluate z 2+ i
(2 x + iy + 1) dz along
1− i
(a) the straight line joining (1 – i) to (2 + i) ts
(b) the curve x = t + 1, y = 2t2 – 1.
z
n
10. Evaluate the line integral (3 y2 dx + 2 y dy) where C is the circle x2 + y2 = 1 counter clockwise
ra
C
from (1, 0) to (0, 1). Y
z
z FGH IJK
pi
11. Evaluate the integral I = dz where C is the

z C C2
boundary of the half annulus as given in figure 1.
z
as
C1
2
12. Evaluate the integral z dz where C is the arc of
C
g
π
the circle |z| = 2 from θ = 0 to θ = . –3 –2 O 2 3 X
.c
3 A B C D
13. Evaluate
z 2z + 3
dz where C is
Fig. 1
w
C z
(i) upper half of the circle |z| = 2 in clockwise direction.
w
(ii) lower half of the circle |z| = 2 in anticlockwise direction.

(iii) the circle |z| = 2 in anticlockwise direction.
z
w
14. Evaluate the integral Re ( z2 ) dz from 0 to 2 + 4i along the line segment joining the points
C
(0, 0) and (2, 4).
15. Evaluate
z0
3+i
( z)2 dz along the real axis from z = 0 to z = 3 and then along a line parallel to
imaginary axis from z = 3 to z = 3 + i. (G.B.T.U. 2013)

16. Integrate f(z) = Re (z) from z = 0 to z = 1 + 2i.
(i) along straight line joining z = 0 to z = 1 + 2i.
(ii) along the real axis from z = 0 to z = 1 and then along a line parallel to imaginary axis from
z = 1 to z = 1 + 2i. (U.P.T.U. 2014)
www.cgaspirants.com
www.cgaspirants.com
Answers
26 26 26 3 i
1. (a) 6 + i (b) 6 + i (c) 6 + i 2. +
3 3 3 2 6
8
3. 10 – i 4. (i) – 1 + i
3
5. (a) (i) 256.5 (ii) 200.25 (b) (i) 26/3 (ii) 26/3 ; No
1 2 2
6. (i) 1 – i (ii) (64i – 103) 7. (i) ;– (ii) 66, – 66
6 3 3
25 4
9. (a) 4 + 8i (b) 4 + i 10. – 1 11.
3 3
− 16
12. 13. (i) 8 – 3πi (ii) 8 + 3πi (iii) 6πi
m
3
26 1 + 2i 1
14. – 8 (1 + 2i) 15. 12 + i 16. (i) (ii) + 2i.
o
3 2 2
.c
1.20 SIMPLY AND MULTIPLY CONNECTED DOMAINS
ts
A domain in which every closed curve can be shrunk to a point without passing out of the
region is called a simply connected domain. If a domain is not simply connected, then it is
called multiply connected domain.
n
ra
1.21 SIMPLY AND MULTIPLY CONNECTED REGIONS
A curve is called simple closed curve if it does not cross itself (Fig. 1). A curve which crosses
pi
itself is called a multiple curve (Fig. 2).

A region is called simply connected if every closed curve in the region encloses points of
as
the region only, i.e., every closed curve lying in it can be contracted indefinitely without pass-
ing out of it. A region which is not simply connected is called a multiply connected region. In
plain terms, a simply connected region is one which has no holes. Figure 3 shows a multiply
g
connected region R enclosed between two separate curves C1 and C2. (There can be more than
.c
two separate curves). We can convert a multiply connected region into a simply connected one,
by giving it one or more cuts (e.g. along the dotted line AB).
w
w
R2 R
w
B
R C
R1 C2
C
C1
Fig. 1 Fig. 2 Fig. 3
Remark 1. Jordan arc is a continuous arc without multiple points.

Remark 2. Contour is a Jordan curve consisting of continuous chain of a finite number of regular arcs.
www.cgaspirants.com
www.cgaspirants.com
1.22 CAUCHY'S INTEGRAL THEOREM [M.T.U. 2013, 2014; G.B.T.U. (C.O.) 2011]
Statement. If f(z) is an analytic function and f ′(z) is continuous at each point within and on a
simple closed curve C, then
z C
f ( z) dz = 0 .
C
Proof. Let R be the region bounded by the curve C. R
Let f(z) = u (x, y) + iv(x, y), then
z f ( z) dz = z (u + iv) (dx + idy)
z z
C C
m
= (udx − vdy) + i (vdx + udy) ...(1)
C C
o
∂u ∂u ∂v ∂v
Since f ′(z) is continuous, the partial derivatives , , , are also continuous
.c
∂x ∂y ∂x ∂y
in R. Hence by Green’s Theorem, we have
zC
f ( z) dz = zz FGH
n R
−
∂v ∂u
−
∂x ∂y
IJ
dx dy + i
K ts zz FGH
R
∂u ∂v
−
∂x ∂y
IJ
dx dy
K ...(2)
Now f(z) being analytic at each point of the region R, by Cauchy-Riemann equations, we
ra
have
∂u ∂v ∂u ∂v
= and =−
∂x ∂y ∂y ∂x
pi
Thus, the two double integrals in (2) vanish.
z
as
Hence f ( z) dz = 0.
C
However Cauchy with the help of Goursat developed the revised form of Cauchy’s
g
fundamental theorem which states that
z
.c
“If f(z) is analytic and one valued within and on a simple closed contour C then f ( z) dz = 0.”
C
w
Goursat showed that for the truth of the original theorem, the assumption of continuity
of f ′(z) is unnecessary and Cauchy’s theorem holds if f(z) is analytic within and on C.
w
Corollary. If f(z) is analytic in a region R and P, Q are two

points in R, then z Q
f ( z) dz is independent of the path joining P Q
w
R
P
and Q and lying entirely in R.
B
Let PAQ and PBQ be any two paths joining P and Q.
By Cauchy’s theorem,
z
A
P
f ( z) dz = 0
z z
PAQBP
⇒ f ( z) dz + f ( z) dz = 0
z z
PAQ QBP
⇒ f ( z) dz − f ( z) dz = 0
z z
PAQ PBQ
Hence f ( z) dz = f ( z) dz.
PAQ PBQ
www.cgaspirants.com
www.cgaspirants.com
1.23 EXTENSION OF CAUCHY'S THEOREM TO MULTIPLY CONNECTED REGION
If f(z) is analytic in the region R between two simple closed

curves C1 and C2, then
z C1
f(z) dz = z
C2
f(z) dz
A B
C2
C1
when integral along each curve is taken in anti-clockwise
direction.
Proof. z f ( z) dz = 0
m
where the path of integration is along AB and curve C2 in clockwise direction and along BA
and along C1 in anti-clockwise direction.
z z z z
o
f ( z) dz + f ( z) dz + f ( z) dz + f ( z) dz = 0
.c
AB C2 BA C1
or zC2
f ( z) dz + z C1
f ( z) dz = 0
n ts ∵ z
AB
f ( z) dz = − zBA
f ( z) dz
Reversing the direction of the integral around C2, we get
z z
ra
f ( z) dz = f ( z) dz
C1 C2
pi
However if a closed curve C contains non-intersecting closed curves C1, C2, ....., Cn,
then by introducing cross-cuts, it can be shown that
z z z z
as
f ( z) dz = f ( z) dz + f ( z) dz + + f ( z) dz .
C C1 C2 Cn
g
C
.c
w
w
C1 C6
w
C5
C2
C4
C3
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Evaluate
Sol.
z C
(x 2 − y 2 + 2ixy) dz , where C is the contour | z | = 1.
f(z) = x2 – y2 + 2ixy = (x + iy)2 = z2 is analytic everywhere within and on | z | = 1.
∴ By Cauchy’s integral theorem, z
C
f ( z) dz = 0.
Example 2. Evaluate zC
y = a(1 – cos θ) between (0, 0) and (2πa, 0).
(3z 2 + 4z + 1) dz where C is the arc of the cycloid x = a(θ – sin θ),
Sol. Here, f(z) = 3z2 + 4z + 1 is analytic everywhere so that the integral is independent
m
of the path of integration and depends only on the end points z1 = 0 + i0 and z2 = 2πa + i0.
z z L O 2 πa
2 πa
2
(3 z + 4 z + 1) dz = (3 z 2
+ 4 z + 1) dz = M z 3
+ 2z 2
+ zP
o
∴ = 2πa (4π2a2 + 4πa + 1).
C 0 N Q 0
.c
Example 3. Evaluate: z 2z + 5 2
dz , where C is the square with vertices at
1 + i, 2 + i, 2 + 2i, 1 + 2i.
C (z + 2) 3 (z 2 + 4)
n
2z2 + 5
ts
ra
Sol. Here, f(z) = Y
( z + 2) 3 ( z 2 + 4)
(2 + 2i)
Singularities are given by (1 + 2i)
pi
D E
(z + 2)3 (z2 + 4) = 0
z = – 2 (order 3), ± 2i (simple poles)
as
C
Since the singularities donot lie inside the contour
C hence by Cauchy’s integral theorem,
z
g
A B
2z2 + 5 (1 + i) (2 + i)
dz = 0.
.c
C ( z + 2) 3 ( z 2 + 4)
z O X
w
Example 4. Evaluate (5z 4 − z 3 + 2) dz around

C
w
(i) unit circle |z| = 1

(ii) square with vertices (0, 0), (1, 0), (1, 1), (1, 0)
w
(iii) curve consisting of the parabola y = x2 from (0, 0) to (1, 1) and y2 = x from (1, 1) to
(0, 0).
Sol. f(z) = 5z4 – z3 + 2 is analytic everywhere. So by Cauchy integral theorem,
z
C
f ( z) dz = 0
∴ For all given curves, z C

(5 z 4 − z 3 + 2) dz = 0.
Example 5. Verify Cauchy theorem by integrating eiz along the boundary of the triangle
with the vertices at the points 1 + i, – 1 + i and – 1 – i. [G.B.T.U. 2012, 2013; M.T.U. 2012)
www.cgaspirants.com
www.cgaspirants.com
Sol. The boundary of triangle C consists of three lines C1, C2 and C3. So,
I= z
C
e iz dz
= z
C1
e iz dz + z C2
e iz dz + zC3
e iz dz = I1 + I2 + I3 ...(1)
Along C1: AB: y = 1, z = x + iy = x + i

∴ dz = dx
I1 = z
C1
e iz dz =
z1
−1
e i( x + i) dx
m
−1
= e (ix − 1) dx
1
o
1 e ix F I −1
e − i−1 − ei−1
= GH JK =
.c
e i 1
i
Along C2 : BE:
∴
x=–1
z = x + iy = – 1 + iy
dz = i dy
n ts
z z
ra
−1
I2 = e iz dz = e i( −1+ iy) i dy
C2 1
pi
= ie
−i
z1
−1
e − y dy = i e–i − e − y e j −1
1
as
1 − i+ 1
= – i (e–i+1 – e–1–i) =
i
e − e −1 − i e j
g
Along C3 : EA: y = x, z = x + iy = (1 + i) x
.c
∴ dz = (1 + i) dx
z z
w
1
I3 = e iz dz = e i(1+ i) x (1 + i) dx
C3 −1
w
L e OP
= (1 + i) M
i(1+ i) x 1
Fe i− 1
− e − i+ 1 I
GH JK
w
=
N i(1 + i) Q −1
i
1 − i−1
From (1), I = I1 + I2 + I3 = e − e i − 1 + e − i + 1 − e −1− i + e i − 1 − e − i + 1 = 0
i
Hence Cauchy’s theorem is verified.
Example 6. Can the Cauchy-integral theorem be applied for evaluating the following
integrals? Hence evaluate these integrals.
(i) z
C
2
e sin z dz; C ≡ |z| = 1 (ii) z
C
tan z dz; C ≡ |z| = 1
(iii) z
C
ez
z2 + 9
dz; C ≡ |z| = 2
www.cgaspirants.com
www.cgaspirants.com
Sol. (i) Let I=

2
z
C
2
e sin z dz
sin z
The integrand f(z) = e is analytic for all z and f ′(z) is continuous inside C. Hence,
Cauchy integral theorem can be applied.
∴ I=0
(ii) Let I= z
C
tan z dz
sin z π
The integrand f(z) = tan z = is analytic for all z except at the points z = ± ,
cos z 2
3π
, ...... . All these points lie outside C. Also f ′(z) is continuous inside C. Hence Cauchy
m
±
2
integral theorem is applicable.
o
∴ I=0
z
.c
ez
(iii) Let I= dz
C z2 + 9
The integrand f(z) =

z2 + 9
ez
ts
is analytic everywhere except at the points z = ± 3i. These
n
points lie outside c and f ′(z) is continuous inside C. Hence Cauchy integral theorem is applica-
ra
ble and I = 0.
ASSIGNMENT
pi
1. (i) State Cauchy-integral theorem for an analytic function. Verify this theorem by integrating
as
the function z3 + iz along the boundary of the rectangle with vertices +1, –1, i, – i.
(U.P.T.U. 2015)
g
(ii) Verify Cauchy’s integral theorem for f (z) = z2 taken over the boundary of a square with vertices
at ± 1 ± i in counter-clockwise direction.
z
.c
2. Using Cauchy’s integral theorem, evaluate f ( z) dz , where f(z) is

C
w
(i) ez (ii) sin z (iii) cos z

(iv) zn ; n = 0, 1, 2, 3, ...... and C is any simple closed path.
w
3. Evaluate: (i)
z z2 − z + 1
dz ; C ≡ |z – 1| =
1
(ii)
z 1
dz ; C ≡ 1 < |z | < 2
w
C z−2 2 C z2 ( z2 + 9)
4. (i) Verify Cauchy’s theorem for f (z) = z3 taken over the boundary of the rectangle with vertices at
– 1, 1, 1 + i, – 1 + i.
(ii) Verify Cauchy’s theorem by integrating z3 along the boundary of a square with vertices at
1 + i, 1 – i, –1 + i and –1 – i. (U.P.T.U. 2014)
5. Evaluate:
(i)
zC
e− z
z+1
dz , where C is the circle |z| =
1
2
(ii)
zC
z2 + 5
z−3
dz , where C is the circle |z| = 1
(iii)
zC
3 z2 + 7 z + 1
z+1
dz , where C is the circle |z + i| = 1
www.cgaspirants.com
www.cgaspirants.com
6. Evaluate the following integrals:
(i)
z
C
z3 + z + 1
z2 − 3 z + 2
dz , where C is the ellipse 4x2 + 9y2 = 1
(ii)
z
C
z+4
z2 + 2 z + 5
dz , where C is the circle |z + 1| = 1
(iii)
z
z2 − z + 1
z−1
dz , where C is the circle |z| =
1
2
z
C
dz
7. Evaluate I = around a triangle with vertices at (0, 0), (1, 0) and (0, 1).
C z−2
8. State and prove Cauchy’s integral theorem. Hence evaluate

z z 2 + 5z + 6
dz where C: |z|=
3
m
C z − 2 2
[M.T.U. 2014, G.B.T.U. (C.O.) 2011]
z e3iz
o
9. Evaluate dz , where C is the circle |z – π| = 3.2. (U.P.T.U. 2007)
( z + π) 3
C
.c
10. (i) Verify Cauchy’s theorem for the function f(z) = 3z2 + iz – 4 along the perimeter of square with
vertices 1 ± i, –1 ± i. (G.B.T.U. 2011)
ts
(ii) Verify Cauchy’s theorem for the function f(z) = 4z2 + iz – 3 along the positively oriented
square with vertices (1, 0), (– 1, 0), (0, 1) and (0, – 1). (M.T.U. 2012)
(iii) Verify Cauchy’s theorem for f(z) = z2 + 3z + 2 where c is the perimeter of square with vertices
n
1 ± i, – 1 ± i. (G.B.T.U. 2012)
ra
Answers
2. 0 in all cases 3. (i) 0 (ii) 0
pi
5. (i) 0 (ii) 0 (iii) 0

6. (i) 0 (ii) 0 (iii) 0
as
7. 0 8. 0 9. 0.
1.24 CAUCHY’S INTEGRAL FORMULA

g
(M.T.U. 2012, U.P.T.U. 2006, 2007, 2009, 2014; G.B.T.U. 2011, 2013)
.c
Statement. If f(z) is analytic within and on a closed curve C and a is any point within C, then
w
1
z f ( z)
w
f(a) = dz .
2πi C z−a
w
f ( z)
Proof. Consider the function , which is analytic at every
z−a C
point within C except at z = a. Draw a circle C1 with a as centre
f ( z) C1
and radius ρ such that C1 lies entirely inside C. Thus is
z−a
analytic in the region between C and C1. aρ
∴ By Cauchy’s theorem, we have
z C
f ( z)
z−a
dz = z
C1
f ( z)
z−a
dz ...(1)
www.cgaspirants.com
www.cgaspirants.com
Now, the equation of circle C1 is

| z – a | = ρ or z – a = ρeiθ
so that dz = iρeiθ dθ
∴ zC1
f ( z)
z−a
dz = z 0
2π f (a + ρe iθ )
ρe iθ
. iρe iθ dθ = i z
0
2π
f (a + ρe iθ ) dθ
Hence by (1), we have

f ( z)
C z−a
dz = i z
2π
0
f (a + ρe iθ ) dθ
In the limiting form, as the circle C1 shrinks to the point a, i.e.,

z ...(2)
ρ → 0, then from (2),
z f ( z)
z−a
dz = i z 2π
f (a) dθ = if (a) z 2π
dθ = 2πif (a)
m
C 0 0
Hence f(a) =
1
z f ( z)
dz
o
2πi C z−a
.c
Aliter: About the point z = a, describe a small circle γ of radius r
f ( z)
lying entirely within C. Consider the function
z−a
.
This function is analytic in the region between C and γ.

n
Hence by Cauchy’s theorem for multiply connected region, we
ts a
ra
γ
have
f (z)
dz = zf (z)
dz z C
pi
C z −a γ z −a
⇒ z f (z)
dz − z f (a)
dz = z f ( z ) − f (a)
dz
as
C z−a γ z−a γ z−a
⇒ z f (z)
dz − f ( a ) z dz
= z f ( z ) − f (a)
dz
g
C z−a γ z−a γ z−a
z z z
.c
f (z) f ( z ) − f (a) dz
dz − 2πif ( a ) = dz ∵ = 2 πi
⇒ γz−a
z−a z−a
w
C γ
since|z − a |= r on γ
z z
w
f (z) f ( z ) − f (a)
⇒ dz − 2πif ( a ) = dz
C z−a γ z−a
z
w
|f ( z ) − f ( a )|
≤ |dz |
γ |z − a |
≤
ε
ε
r γ
| dz| z ∵ f ( z) is continuous at z = a
∴ | f ( z) − f (a)|< ε
≤ . 2πr and| z − a|= r for z on γ
r
≤ 2πε → 0 as ε → 0
⇒ z C
f ( z)
z−a
dz – 2πif(a) = 0
⇒ f(a) =
1
2πi z
C
f ( z)
z−a
dz
www.cgaspirants.com
www.cgaspirants.com
1.25 CAUCHY’S INTEGRAL FORMULA FOR THE DERIVATIVE OF AN ANALYTIC

FUNCTION [U.P.T.U. (C.O.) 2009, 2010]
If a function f(z) is analytic in a region D, then its derivative at any point z = a of D is also
analytic in D and is given by
f ′(a) =
1
2πi z C
f(z)
(z − a) 2
dz
where C is any closed contour in D surrounding the point z = a.

Proof. Let a + h be a point in the neighbourhood of the point a. Then by Cauchy’s Integral
Formula
m
1 f ( z)
f(a) = dz
2πi C z−a
o
1 f ( z)
f(a + h) = dz
.c
2πi C z−a−h
∴ f(a + h) – f(a) =
1
2πi
n z RST
C
1
−
1
z−a−h z−a
f ( z) dz =
ts
h
2πi
UV
W z
C
f ( z) dz
( z − a − h) ( z − a)
f (a + h) − f (a) 1
z f ( z) dz
ra
⇒ =
h 2πi C ( z − a − h) ( z − a)
Take limit as h → 0
z
pi
f (a + h) − f (a) 1 f ( z) dz
Lt = Lt
h→0 h h → 0 2πi C ( z − a − h) ( z − a)
as
⇒ f ′(a) =
1
2πi z f ( z)
( z − a) 2
dz ...(1)
g
C
.c
Since a is any point of the region D, so by (1) it is clear that f ′(a) is analytic in D. Thus,
the derivative of an analytic function is also analytic.
w
1.26 THEOREM
w
If a function f(z) is analytic in a domain D, then at any point z = a of D, f(z) has derivatives of
all orders, all of which are again analytic functions in D, their values are given by
w
f n(a) =
n!
2πi z f(z)
C (z − a) n + 1
dz
where C is any closed contour in D surrounding the point z = a.

Proof. We shall prove this theorem by Mathematical Induction.
Let the theorem be true for n = m. Then
f m(a) =
m!
2πi z C
f ( z)
( z − a) m + 1
dz is true.
⇒
f m (a + h) − f m (a) m ! 1
h
=
2 πi h
LM
N zC
f ( z) dz
( z − a − h) m+1
− z
C
f ( z) dz
( z − a) m+1
UV
W
www.cgaspirants.com
www.cgaspirants.com
R|F h I U
=
m! 1
. .
2 πi h zC
1
( z − a) m + 1
S|GH 1 − z − a JK − 1|V| f (z) dz
T
− ( m + 1)
W
=
m! 1
. .
2πi h zC
1
( z − a) m +1
R|S(m + 1) h + (m + 1) (m + 2) h
|T z−a 2! ( z − a)
2
2
+
U|V f (z) dz
|W
Take limit as h → 0
h→0
f m (a + h) − f m (a) (m + 1) !
lim
h
=
2πi
f ( z)
z
C ( z − a) m + 2
dz
⇒ f m+1(a) =
(m + 1) ! f ( z)
z dz
m
2πi C ( z − a) m + 2
Hence the theorem is true for n = m + 1 if the theorem is true for n = m. But we know by
Cauchy’s Integral formula for the derivative of a function that the theorem is true for n = 1.
o
Hence the theorem must be true for n = 2, 3, 4, ...... and so on i.e., for all +ve integral values
.c
of n. Thus,
f n(a) =
n!
2 πi
n z C
f ( z)
( z − a) n + 1
dz
ts ...(1)
Since a is any point of the region D, so by (1) it is clear that f n(a) is analytic in D. Thus
ra
the derivatives of f(z) of all orders are analytic if f(z) is analytic.

Thus, if a function of a complex variable has a first derivative in a simply connected
pi
region, all its higher derivatives exist in that region. This property is not exhibited by the functions
of real variables.
as
1.27 CAUCHY’S INEQUALITY
If f(z) is analytic within a circle C given by | z – a | = R and if | f(z) | ≤ M on C, then

g
.c
Mn !
| f n(a) | ≤ .
Rn
z
w
n! f ( z) dz
Proof. f n(a) = 2πi
( z − a) n + 1
w
n!
z f ( z) dz
w
∵ z – a = R e iθ
⇒ | f n(a) | = n+ 1
2 πi C ( z − a) ∴ dz = i R e iθ dθ
n!
≤ |2πi| z | f ( z)|| dz|
C |( z − a) n + 1 |
∴ | dz|=|i R e iθ dθ|
= R dθ
≤
n! M
2π R n + 1 z0
2π
R dθ
n! M Mn !
≤ n + 1 2πR ≤ .
2π R Rn
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
e− z
z+1
dz , where C is the circle | z | = 2
Sol. f(z) = e–z

is an analytic function.
The point a = – 1 lies inside the circle | z | = 2.
∴ By Cauchy’s integral formula,
z C
e− z
z+1
dz = 2πi(e–z)z= –1 = 2πie.
m
Example 2. Evaluate the following integral:
1
z
o
cos z dz
C z
.c
where C is the ellipse 9x2 + 4y2 = 1.
Sol. Pole is given by z = 0. The given ellipse encloses the simple pole.
z z
cos z
C
n
dz = 2πi (cos z)z=0 = 2πi.
ts
ra
Example 3. (i) Use Cauchy Integral formula to evaluate
z
sin πz 2 + cos πz 2
dz
pi
C (z − 1)(z − 2)
where C is the circle | z | = 3. [G.B.T.U. 2010; G.B.T.U. (C.O.) 2011]
z
as
sin πz + cos πz
(ii) Evaluate: dz, where C is the circle | z | = 4. (U.P.T.U. 2008)
C (z − 1) (z − 2)
g
Sol. (i) The integrand has singularities given by

.c
Y
(z – 1) (z – 2) = 0 ⇒ z = 1, 2
The given circle | z | = 3 with centre at z = 0 and radius C≡|z|=3
w
3 encloses both the singularities.

F sin πz + cos πz I
2 2
w
GH z − 2 JK
z z
z=2
∴ sin πz 2 + cos πz 2 O z = 1
X
dz = dz
w
C (z − 1 )(z − 2 ) C1 z−1
F sin πz + cos πz I
2 2
GH z − 1 JK
+ z
C2 z−2
dz
= 2πi
LM sin πz + cos πz OP + 2πi LM sin πz + cos πz OP
2 2 2 2
N z−2 Q N z−1 Q
z= 1 z=2
F 0 − 1IJ + 2πi FG 0 + 1IJ = 2πi + 2πi = 4πi.

= 2πi G
H −1 K H 1 K
www.cgaspirants.com
www.cgaspirants.com
(ii) Singularities are given by

(z – 1) (z – 2) = 0 ⇒ z = 1, 2
The given circle | z | = 4 with centre at z = 0 and radius 4 encloses both the singularities.
FG sin πz + cos πz IJ FG sin πz + cos πz IJ
∴ zC
sin πz + cos πz
( z − 1) ( z − 2)
dz = z C1
H z − 2 K dz + H z − 1 K dz
z−1 z
C2 z−2
= 2πi
LM sin πz + cos πz OP + 2πi LM sin πz + cos πz OP
N z−2 Q N z−1 Q
z=1 z=2
= 2πi M
L − 1OP + 2πi L 1O = 2πi + 2πi = 4πi
N − 1Q MN 1PQ
m
Example 4. (i) Evaluate the following integral using Cauchy Integral formula
o
4 − 3z
dz , where C is the circle | z | = 3/2.
.c
C z (z − 1)(z − 2)
(U.P.T.U. 2015)
(ii) Use Cauchy-integral formula to evaluate
ts
z
n
z 1
dz, where C is the circle z − 2 = . (U.P.T.U. 2009)
z 2 − 3z + 2
ra
C 2
Sol. (i) Poles of the integrand are z = 0, 1, 2. These are simple poles.
pi
3 3
Given circle | z | = with centre at z = 0 and radius encloses two poles z = 0 and z = 1.
2 2
as
4 − 3z 4 − 3z
∴ z 4 − 3z
dz = z ( z − 1) ( z − 2)
dz + z z ( z − 2)
dz
g
C z ( z − 1) ( z − 2) C1 z C2 ( z − 1)
LM 4 − 3z OP LM 4 − 3z OP
.c
= 2πi + 2πi = 2πi.

N (z − 1) (z − 2) Q N z (z − 2) Q
w
z=0 z= 1
(ii) Poles of the integrand are given by

w
z2 – 3z + 2 = 0 ⇒ z = 1, 2
1 1
w
Both are simple poles. The given circle | z – 2 | = with centre at z = 2 and radius
2 2
encloses only one of the poles at z = 2.
FG z IJ
H z − 1K dz = 2πi LM z OP
zC 2
z
z − 3z + 2
dz = zC z−2 N z − 1Q z=2
= 2πi
FG 2 IJ = 4πi
H 1K
Example 5. Evaluate by Cauchy’s integral formula
zC
dz
z( z + πi)
, where C is | z + 3i | = 1
www.cgaspirants.com
www.cgaspirants.com
Sol. Poles of the integrand are z = 0, – πi (simple poles)

The given curve C is a circle with centre at z = – 3i, i.e., at (0, – 3) and radius 1.
Clearly, only the pole z = – πi lies inside the circle.
FG 1IJ
∴ z
C
dz
z( z + πi)
= z
C
H zK
z + πi
dz
= 2π i FG 1IJ | By Cauchy’s Integral formula

H zK z = − πi
2πi
= =–2
m
− πi
z z2 + 1
o
Example 6. Evaluate dz where C is circle,
C z2 − 1
.c
(i) | z | = 3/2 (ii) | z – 1 | = 1 (iii) | z | = 1/2.
(U.P.T.U. 2014) (U.P.T.U. 2014)
Sol. The integrand has singularities given by
z2 – 1 = 0 ⇒ z = ± 1
n ts
ra
(i) The given curve C is a circle with centre at origin Y
(0, 0) and radius 3/2.
C ≡ | z | = 3/2
Both the singularities z = 1 and z = – 1 lie inside the
pi
circle | z | = 3/2.
F z + 1I
2 F z + 1I 2
as
GH z + 1 JK GH z − 1 JK
z z z
X
∴ z2 + 1 z=–1 O z=1
dz = dz + dz
z2 − 1 z−1 z+1
g
C C1 C2
F z + 1I F z + 1I
.c
2 2
= 2πi GH z + 1 JK + 2πi GH z − 1 JK
w
z= 1 z=−1
| By Cauchy’s Integral formula

w
= 2πi (1) + 2πi (– 1) = 0 Y

(ii) The given curve C is a circle with centre at C ≡ | z – 1| = 1
w
(1, 0) and radius 1.

Only the singularity z = 1 lie inside the given
z=1
circle | z – 1 | = 1. O
X
F z + 1I (0, 0) (1, 0)
2
GH z + 1 JK
∴ zC
2
z +1
z2 − 1
dz = z
C z−1
dz
F z + 1I2
= 2πi GH z + 1 JK z= 1
= 2πi | By Cauchy’s Integral formula
www.cgaspirants.com
www.cgaspirants.com
(iii) The given curve C is a circle with centre at Y

1 C ≡ | z | = 1/2
origin (0, 0) and radius . Clearly both the
2
singularities z = 1 and z = – 1 lie outside the
1 X
given circle | z | =
. O
2
Hence, by Cauchy’s Integral theorem
z
C
z2 + 1
z2 − 1
dz = 0.
m
Example 7.(i) Use Cauchy’s integral formula to show that
z e zt
dz = 2πi sin t if t > 0 and C is the circle |z| = 3. (U.P.T.U. 2009)
o
C z2 + 1
.c
(ii) Evaluate the following complex integration using Cauchy’s integral formula
z
C
3z 2 + z + 1
(z 2 − 1)(z + 3)
n dz
ts
where C is the circle | z | = 2
Sol. (i) Singularities of the integrand are given by

ra
z2 + 1 = 0 ⇒ z = ± i (order 1)
The circle |z| = 3 has centre at z = 0 and radius 3. It encloses both the singularities z = i
and z = – i.
pi
Fe I zt Fe I zt
GH z + i JK GH z − i JK
z z z z
as
e zt e zt
Now, dz = dz = dz + dz
C z2 + 1 C ( z − i) ( z + i) C1 z−i C2 z+i
Fe I Fe I
g
zt zt
= 2πi GH z + i JK + 2πi GH z − i JK = π (eit – e– it) = 2πi sin t
.c
z=i z=−i
w
(ii) Poles of the integrand are given by

(z2 – 1) (z + 3) = 0 ⇒ z = 1, – 1, – 3 (simple poles)
w
The circle | z | = 2 has centre at z = 0 and radius 2. clearly the poles z = 1 and z = – 1
lie inside the given circle while the pole z = – 3 lie outside it.
w
RS 3z + z + 1 UV
2 RS 3z + z + 1 UV 2
∴ zC
2
3z + z + 1
( z 2 − 1)( z + 3)
dz = z
C1
T ( z + 1)(
z−1
z + 3)
z
W dz + T (z − 1)(z + 3) W dz
C2 z+1
= 2πi
LM 3z + z + 1 OP + 2πi LM 3z + z + 1 OP
2 2
N (z + 1)(z + 3) Q z=1N (z − 1)(z + 3) Q z=−1

| Using Cauchy’s Integral formula
F 5I F 3I F − 1IJ = – πi
= 2πi GH JK + 2πi GH − JK = 2πi GH
8 4 8 K 4
www.cgaspirants.com
www.cgaspirants.com
Example 8. Integrate (z 3 – 1) –2 the counterclockwise sense around the circle

|z–1|=1
Y
Sol. Singularities of integrand are given by
C ≡ | z – 1| = 1
(z3 – 1)2 = 0
⇒ (z – 1)2 (z2 + z + 1)2 = 0
− 1± i 3
⇒ z = 1, O
z=1 X
2
Singularities are of second order.
The circle | z – 1 | = 1 has centre at z = 1 and
radius 1. Clearly, only z = 1 lies inside the circle
m
|z–1|=1
RS UV
o
1
z dz
z T (z 2
+ z + 1) 2 W dz
.c
Now, =
C ( z − 1) 2
3
C ( z − 1) 2
LM R UVOP
=
2πi d
n
MN TS 1
1 ! dz ( z + z + 1) 2
2
WPQ ts
Using Cauchy's Integral
formula for derivatives
z=1
L − 2 (2 z + 1) OP = – 4πi FG 3 IJ = – 4πi
ra
= 2πi M
N (z + z + 1) Q 2 H 27 K 93
z=1
z
pi
ez
Example 9. Evaluate: dz, where C is | z | = 4. (U.P.T.U. 2008)
C (z 2 + π 2 ) 2
as
Sol. Singularities of the integrand are given by

(z2 + π2)2 = 0 ⇒ z = ± πi (order 2)
g
The given curve C is a circle with centre at origin and radius 4. The circle encloses both
.c
the singularities.
RS e UV z RS e UV z
w
∴ z ez
dz = z T (z + πi) W dz + T (z − πi) W dz
2
z 2
w
C (z2 + π 2 ) 2 C1 ( z − πi) 2 C2 ( z + πi) 2
LM d R e UOP LM d R e UVOP
w
z z
= 2πi
MN dz ST ( z + πi) VWPQ 2
z = πi
+ 2πi
MN dz ST ( z − πi) 2
WPQ z = − πi
| By C–I formula for derivatives
= 2πi
LM e z
( z + πi − 2) OP + 2πi
LM e z
( z − πi − 2) OP
N ( z + πi) 3 Q z = πi N ( z − πi) 3 Q z = − πi
FG πi − 1IJ + FG πi + 1IJ = i .
=
H 2π K H 2π K π
2 2
www.cgaspirants.com
www.cgaspirants.com
Example 10. Use Cauchy’s integral formula to evaluate
zC
e 2z
(z + 1) 4
dz where C is the circle | z | = 2. [U.K.T.U. 2011]
Sol. The integrand has a singularity at z = – 1 which lies within the circle | z | = 2.
z
C
e2 z
( z + 1) 4
dz =
2πi d 3 2 z
3 ! dz 3
RS
(e )
T
UV
W z =−1
=
πi
3
8πi
(8e2z)z = –1 = 2 .
3e
Example 11. Evaluate z C z +12

z
dz , where
m
1
(i) C ≡ z + =2 (ii) C ≡ | z + i | = 1.
z
o
Sol. Poles of the integrand are given by
z2 + 1 = 0 ⇒
.c
z=±i
Integrand has two simple poles z = i and z = – i
(i) The given curve is
n
z+
1
z
=2
ts Y
C1 ≡ | z – i | = 2
ra
2
1
⇒ x + iy + =2 i
x + iy
pi
O
x 2 – y 2 + 2ixy + 1 X
as
⇒ =2
x + iy
C2 ≡ | z + i | = 2
g
⇒ (x2 – y2 + 1)2 + 4x2y2 = 4x2 + 4y2 –i

.c
⇒ x4 + y4 – 2x2y2 +1+ 2x2 – 2y2 + 4x2y2 2

= 4x2 + 4y2
w
⇒ (x2 + y2 ) 2 – 2(x2 + y2 ) +1= 4y2

w
⇒ x2 + y2 – 1 = ± 2y
⇒ x2 + (y ± 1)2 = 2
w
Above eqn. represents two circles with centres (0, 1), (0, – 1) and radius 2.
z C 2
z +1
z
dz = zC1 z +12
z
dz + z C2 2
z +1
z
dz
FG z IJ FG z IJ
= zH C1
z+i
z−i
dz +
K
zC2
H z − i K dz
z+i
= 2πi
FG z IJ + 2πi
FG z IJ
H z + iK z=i
H z + iK z=−i
www.cgaspirants.com
www.cgaspirants.com
FG 1IJ + 2πi FG 1IJ = 2πi.

= 2πi
H 2K H 2K Y
(ii) The given curve | z + i | = 1 is a circle with
centre at z = – i and radius 1. Clearly only the pole
O (0, 0)
z = – i lies inside the circle | z + i | = 1 X
FG z IJ C≡|z+i|=1
zC
z
z2 + 1
dz = z C
H z − i K dz
z+i (0, –1) z=–i
= 2πi
FG z IJ
H z − iK
m
z=−i
= πi | By Cauchy Integral formula
o
Example 12. Evaluate by using Cauchy Integral formula
z
.c
z−1
dz , where C is |z – i| = 2.
C ( z + 1) 2 ( z − 2)
Sol. Poles of the integrand are given by
⇒
(z + 1)2 (z – 2) = 0
z = – 1, 2
n ts Y
C≡|z–i|=2
ra
(0, 3)
z = – 1 is a double pole while z = 2 is a simple
pole.
pi
(0, 1)
The given curve C is a circle with centre at (0, 1)
2
and radius 2. Clearly, the pole z = – 1 lies inside the z=–1
1
z=2
given circle while the pole z = 2 lies outside it.
as
(–1, 0) O (0, 0) X
Hence, ( 3, 0)
FG z − 1IJ (0, –1)

g
z z−1
z H z − 2 K dz
.c
dz =
C ( z + 1) 2 ( z − 2) C ( z + 1) 2
RS FG IJ UV
w
2πi d z − 1
=
T H
1 ! dz z − 2 KW
w
z=−1
R −1
= 2πi S
UV =–
2πi
.
w
T (z − 2) 2
W z =−1
9
ASSIGNMENT
1. Evaluate
zC
z2 + 5
z−3
dz , where C is the circle | z | = 4.
2. Evaluate
zC
2
ez
z +1
dz over the circular path | z | = 2.
3. Evaluate
zC
3 z2 + 7 z + 1
z+1
dz , where C is the circle | z | = 1.5.
www.cgaspirants.com
www.cgaspirants.com
4.
5.
Evaluate
cos z
z−π zdz , where C is the circle | z – 1 | = 3.
C
Evaluate the complex integration
(i)
zC
|RS cos πz + sin πz U|V dz where C is the circle | z | = 3.
2
T| ( z + 1)( z + 2) |W
2
(ii)
zC
sin πz2 + cos πz2
( z − 1) ( z − 3)
dz where C: |z| = 2 (M.T.U. 2013)
6. (i) Evaluate zz dz
( z − 1)( z − 3)
(a) | z | = 3
C
, where C is the circle
(b) | z | = 3/2.
m
ez
(ii) Evaluate dz , where C is the circle | z | = 2.
C ( z − 1)( z − 4)
o
(iii) Evaluate using Cauchy’s integral formula:
e2 z
z
.c
dz where C is the circle | z | = 3.
C ( z − 1)( z − 2)
ts
(iv) State Cauchy’s integral formula. Hence evaluate: (G.B.T.U. 2011, 2012)
z exp ( iπz )
dz
n
C ( 2z 2 − 5z + 2 )
ra
where C is the unit circle with centre at origin and having positive orientation.
7. (i) Evaluate
z ez
z( z + 1)
1
dz , where C is the circle | z | = .
4
pi
(ii) Using Cauchy Integral formula, evaluate

z dz
where C ≡ | z | = 2.
as
C z2 − 1
(iii) Evaluate z 2z + 1
dz where C is | z| =
1
.
g
2 2
C z +z
z
.c
cos πz
8. Evaluate dz around a rectangle with vertices
z2 − 1
C
w
(a) 2 ± i, – 2 ± i (b) – i, 2 – i, 2 + i, i.
w
ez
9. Integrate 2
around the contour C, where C is
z +1
w
(i) | z – i | = 1 (ii) | z + i | = 1
10. Show that

z
C
ez
z
dz = 2πi, C ≡ | z | = 1. Hence show that
z0
2π
ecos θ cos (sin θ) dθ = 2π and
z 2π
0
ecos θ sin (sin θ) dθ = 0
11. Evaluate
z
z2 + 9
C
(i) | z – 3i | = 4
dz
, where C is
(ii) | z + 3i | = 2 (iii) | z | = 5
www.cgaspirants.com
www.cgaspirants.com
12. Evaluate:
(i) z
C
2
z+4
z + 2z + 5
dz ; C ≡ | z + 1 – i | = 2 (ii)
z
C
z3 − 6
2z − i
dz ; C ≡ | z | = 1
(iii)
z
C
tan z
z2 − 1
dz ; C ≡ | z | = 3/2 (iv) z C
2 z2 + z
z2 − 1
dz ; C ≡ | z – 1 | = 1.
13. Evaluate by Cauchy-Integral formula:

zC
z2 + 1
z2 − 1
dz, where C is
(i) | z – 1 | = 1 (ii) | z + 1 | = 1 (iii) | z – i | = 1.

14. Evaluate the following integrals:
z cos 2πz
z z 4 − 3 z2 + 6
m
(i) dz; C ≡ | z | = 1 (ii) dz; C ≡ | z | = 2
C (2 z − 1) ( z − 3) C ( z + i)3
z cosh z
o
(iii) dz; C ≡ | z | = 1/2
C z4
z FGH
.c
sin 2 z
15. Evaluate dz , where C is the circle | z | = 1.
C π IJ 3
16. (i) Evaluate

z
z−
C
6 K
e−2 z dz
( z + 1)3
n
, where C is the circle | z | = 2.
ts
ra
(ii) Evaluate the integral

z e2z
( z + 1) 5
dz around the boundary of the circle |z| = 2.
pi
(U.P.T.U. 2015)
17. (i) Evaluate:

z e 3z
dz, where C is the square with vertices at ± 1 ± i.
as
C ( z − log 2)4
(ii) Evaluate:
z dz
, where C ≡ | z | = 1. (G.B.T.U. 2013)
g
C z 2 ( z 2 − 4) e z
z
.c
ez
18. Evaluate dz, where C is
C z (1 − z)3
w
1 1
(i) | z | = (ii) | z – 1 | = (iii) | z | = 2
2 2
w
sin 2 z
19. Integrate around the contour C, where C is a rectangle with vertices at 3 ± i, – 2 ± i.
( z + 3) ( z + 1)2
w
20. Evaluate:
(i) | z | = 3
z C
z3 − z
( z − 2)3
dz, where C is
(ii) | z – 2 | = 1 (iii) | z | = 1
21. Using Cauchy-integral formula, evaluate:
(i)
z
C
cos z
( z − πi)2
dz; C≡|z|=5 (ii)
z
C
ez
z3
dz ; C≡|z|=1
(iii) z
C
ez
( z − 1) ( z2 + 4)
2 dz; C≡|z–1|=
1
2
(iv)
z
C
e zt
( z 2 + 1) 2
dz; C ≡ | z | = 3, t > 0.
www.cgaspirants.com
www.cgaspirants.com
22. Evaluate
z
sin z
z2 − iz + 2
(i) | z + 2 | = 2
C
dz, where C is
(ii) A rectangle with vertices at (1, 0), (1, 3), (– 1, 3) and (– 1, 0)

(iii) A rectangle with vertices at (2, 0), (2, 3), (– 2, 3) and (– 2, – 3).
23. Evaluate the integrals
24.
(i)
z e z + sin πz
( z − 1) ( z − 3)2 ( z + 4)
C
Show that
dz, C≡|z|=2 (ii)
z
C
z+1
z2 − 9
dz; C ≡ | z + 3 | = 1
(i)
z dz
=
π
; C≡|z–i|=2
m
2 2 16
C ( z + 4)
(ii) z ez
dz =
FG 11 − 4IJ πi ; C ≡ | z | = 2
He K
o
2 3
C z ( z + 1)
z dz
.c
(iii) = 0; C ≡ | z – 1 | = 4
C ( z 2 + 4) 3
25. Evaluate
z C
z
( z − 6 z + 25)2
2
ts
dz by Cauchy integral formula, where C is | z – 3 – 4i | = 4.
z z z
n
P( z ) P( z ) P( z )
26. Let P(z) = a + bz + cz2 and dz = dz = dz = 2πi where C is the circle | z | = 1.
ra
2
z C z C z3
Evaluate P(z).
z
pi
3z 2 + 7z + 1
27. If f(ξ) = dz, where C is the circle x2 + y2 = 4, find the values of f(3), f ′(1 – i) and
C z−ξ
as
f ″(1 – i).
28. Evaluate:
zC
(1 + z) sin z
(2 z − 3) 2
dz, where C ≡ | z – i | = 2 counter-clockwise. (U.P.T.U. 2014)
g
.c
Answers
1. 28πi 2. 2πi sin 1 3. – 6πi 4. – 2πi
w
5. (i) – 4πi (ii) πi

w
2 2π
6. (i) (a) 2πi (b) – πi (ii) – πie (iii) 2πi (e4 – e2) (iv)
3 3
w
7. (i) 2πi (ii) 0 (iii) 2πi

8. (a) 0 (b) – πi
9. (i) π(cos 1 + i sin 1) (ii) – π(cos 1 – i sin 1)
π π
11. (i) (ii) – (iii) 0
3 3
π π
12. (i) (3 + 2i) (ii) – 6πi (iii) 2πi tan 1 (iv) 3πi
2 8
13. (i) 2πi (ii) – 2πi (iii) 0
2πi
14. (i) (ii) – 18πi (iii) 0 15. πi
5
4 πi
16. (i) 4πie2 (ii) 2
3e
www.cgaspirants.com
www.cgaspirants.com
πi
17. (i) 72πi (ii) –
2
18. (i) 2πi (ii) – πie (iii) πi (2 – e)
πi
19. (4 cos 2 + sin 2) 20. (i) 12πi (ii) 12πi (iii) 0
2
6πe
21. (i) 2π sinh π (ii) πi (iii) (iv) πi (sin t – t cos t)
25
2πi 2πi
22. (i) 0 (ii) sinh 2 (iii) (sinh 2 + sinh 1)
3 3
πie 2πi 3π
23. (i) (ii) 25. .
10 3 128
m
26. P(z) = 1 + z + z2 27. f(3) = 0, f ′(1 – i) = 2π(6 + 13i), f ″(1 – i) = 12πi
πi 5 FG3 3 IJ
o
28. cos + sin
2 2 H2 2 K
.c
1.28 REPRESENTATION OF A FUNCTION BY POWER SERIES
A series of the form

∞
∑a z n
n
or
∞
∑a
n
n (z
ts
− a) n whose terms are variable is called a power
n=0
ra
n=0
series, where z is a complex variable and an, a are complex constants. The second form can be
reduced to first form merely by substitution z = ζ + a or by changing the origin.
pi
Every complex function f(z) which is analytic in a domain D can be represented by a

power series valid in some circular region R about a point z0. Both the circular region R and
the point z0 lie inside D. Such a power series is Taylor’s series. If f(z) is not analytic at a point
as
z0, we can still expand f(z) in an infinite series having both positive and negative powers of
z – z0. This series is called the Laurent’s series.
g
1.29 TAYLOR’S SERIES [U.P.T.U. (C.O.) 2008]

.c
If f(z) is analytic inside a circle C with centre at a, then for all z inside C,
w
(z − a) 2 (z − a)n n
f(z) = f(a) + (z – a) f ′(a) + f ″ (a) + + f (a) + ⋅⋅⋅
2! n!
w
Or
w
∞
f (n) (a)
f(z) = ∑
n =0
an (z − a) n , where an =
n!
.
Proof. Let z be any point inside the circle C. Draw a circle C1 with C
centre at a and radius smaller than that of C such that z is an w
interior point of C1. Let w be any point on C1, then
z−a
|z–a|<|w–a| i.e., <1 z
w− a C1 a
Now,
1
=
1
=
1
1−
z−a LM OP −1
w − z (w − a) − ( z − a) w − a w− a N Q
www.cgaspirants.com
www.cgaspirants.com
z−a
Expanding RHS by binomial theorem as < 1, we get
w− a
1 1 z−aLM z−a FG IJ 2
F z − a IJ n OP
= 1+ + + +G + ...(1)
w− z w− a w− aMN w− a H K H w − aK PQ
z−a
This series converges uniformly since < 1. Multiplying both sides of eqn. (1) by
w− a
1
f (w) and integrating term by term w.r.t. w, over C1, we get
2πi
1
z f (w)
dw =
1 f (w)
z
dw +
z−a f (w)
dw + z
( z − a) 2
z f (w)
dw
m
2πi C1 w − z 2πi C1 w − a 2πi C1 (w − a) 2 2πi C 1 (w − a) 3
+ ⋅⋅⋅ +
( z − a) n
z f (w)
dw + ⋅⋅⋅ ...(2)
o
2πi C1 (w − a) n + 1
.c
( z − a) 2 ( z − a) n n
⇒ f(z) = f(a) + (z – a) f ′(a) + f ″ (a) + ⋅ ⋅ ⋅ + f (a) + ⋅⋅⋅ ...(3)
which is the required Taylor’s series for f(z) about z = a.

Cor. 1. Putting z = a + h in (3), we get
n 2!
ts n!
ra
h2 hn n
f(a + h) = f(a) + hf ′(a) + f ″(a) + ⋅⋅⋅ + f (a) + ⋅⋅⋅
2! n!
pi
Cor. 2. If a = 0, the series (3) becomes

z2 zn n
as
f(z) = f(0) + zf ′(0) + f ″(0) + ⋅⋅⋅ + f (0) + ⋅⋅⋅

2! n!
This series is called Maclaurin’s series.
g
1.30 LAURENT’S SERIES [U.P.T.U. (C.O.) 2008]

.c
If f(z) is analytic inside and on the boundary of the annular (ring shaped) region R bounded by
w
two concentric circles C1 and C2 of radii r1 and r2 (r1 > r2) respectively having centre at a, then
for all z in R,
w
∞ ∞
f(z) = ∑ an (z – a)n + ∑ b (z − a) −n
w
n
n=0 n=1
where, an =
1
2πi z C1
f(w)
(w − a)n +1
dw ; n = 0, 1, 2, ...
and bn =
1
2πi z
(w − a) − n +1
C2
f(w)
dw ; n = 1, 2, 3, ...
Proof. Let z be any point in the region R, then by Cauchy’s integral formula for double connected
region, we have
f(z) =
1
2πi 1 w − z
C z
f (w)
dw −
1 f (w)
2πi 2 w − z
C
dw z ...(1)
www.cgaspirants.com
www.cgaspirants.com
For the first integral in (1), w lies on C1

C1
z−a C2
∴ |z–a|<|w–a| i.e., <1
w− a
z
Now
1
=
1
=
1
1−
z−a FG IJ −1 r1 a
r2
w − z (w − a) − ( z − a) w − a w− a H K
LM FG IJ 2 O R
+ ⋅ ⋅ ⋅P
1 z−a z−a
= 1+ +
w− a w− a MN
w− a H K PQ
1
Multiplying both sides by f (w) and integrating term by term w.r.t. w, along the
2πi
m
circle C1, we get
1
z f (w) 1
z f (w) z−a
z f (w) ( z − a) 2
z f (w)
o
dw = dw + dw + dw
2πi C1 w − z 2πi C1 w − a 2πi C1 (w − a) 2 2πi C1 (w − a) 3
.c
+ ⋅⋅⋅
= a0 + a1(z – a) + a2(z – a)2 + ⋅⋅⋅
= ∑a
∞
n=0
n (z
n − a) n ...(2)
LM∵
MN an =ts 1
2πi z
C1
f (w)
(w − a) n +1
dw, n = 0, 1, 2, ⋅ ⋅ ⋅
OP
PQ
ra
For the second integral in (1), w lies on C2
w− a
∴ | w – a | <| z – a | i.e., <1
pi
z−a
Now
1 1 1 w−a FG IJ −1
as
= =− 1−
w − z (w − a) − ( z − a) z−a z−a H K
1 w− a LM
w− a FG IJ 2 OP
g
=– 1+ + + ⋅⋅⋅
z−a z−a MN
z−a H K PQ
.c
1
w
Multiplying both sides by – f (w) and integrating term by term w.r.t. w, along the
2πi
circle C2, we get
w
–
1
z f (w)
dw =
1
⋅
1
z f (w) dw +
1
⋅
1
z (w – a) f(w) dw
w
2
2πi C2 w− z z − a 2πi C2 ( z − a) 2πi C2
+
( z − a)
1
3
⋅
1
2πi z
C2
(w – a)2 f(w) dw + ⋅⋅⋅
= b1(z – a)–1 + b2(z – a)–2 + b3(z – a)–3 + ⋅⋅⋅
=
∞
∑ b ( z − a)
n=1
n
−n
...(3)
LM∵
MN bn =
1
2πi zC2
f (w)
(w − a) − n +1
dw, n = 1 , 2, 3, ⋅ ⋅ ⋅
OP
PQ
Substituting from (2) and (3) in (1), we get
∞ ∞
f(z) = ∑
n=0
an ( z − a) n + ∑ b ( z − a)
n=1
n
−n
www.cgaspirants.com
www.cgaspirants.com
Note 1. In case f(z) is analytic inside C1, then bn = 0 and an =

1
2πi z
C1
f (w)
(w − a) n+1
dw =
f n(a)
n!
and Laurent’s series reduces to Taylor’s series.
Note 2. If C is any simple closed curve which lies in the ring-shaped region R and encloses the circle C1,
then
z C1
f (w)
(w − a) n+1
dw =
z
C
f (w)
(w − a) n + 1
dw
and z C2
f (w)
(w − a) − n+1
dw =
z
C
f (w)
(w − a) − n + 1
dw
m
∴ Laurent’s series can be written as
z
C
∞
1 f (w)
f(z) = ∑ an ( z − a) n , where an = dw .
o
2πi C (w − a) n + 1
n=−∞
.c
Example 1. Expand
1
n EXAMPLES
in the region
ts
z 2 − 3z + 2
ra
(a) | z | < 1 (b) 1 < | z | < 2 (U.P.T.U. 2015)

(c) | z | > 2 (d) 0 < | z – 1 | < 1. (G.B.T.U. 2006, 2008, 2010)
pi
1 1 1 1
Sol. Here f(z) = = = − | Partial Fractions
z2 − 3z + 2 ( z − 1)( z − 2) z − 2 z − 1
as
(a) When | z | < 1

1 1
g
f(z) = +
∴
FG z IJ 1− z
.c
− 2 1−
H 2 K
[Arranged suitably to make the binomial expansions valid]
w
1 FG
z IJ −1
1
∑ FGH 2 IJK + ∑
∞
z
n ∞
w
zn
=–
2
1−
2H K + (1 – z)–1 = –
2 n=0 n=0
w
This is a series in positive powers of z, so it is an expansion of f(z) in Taylor’s series

within the circle | z | = 1.
(b) When 1 < |z| < 2
1 1 1 FGz IJ −1
1FG 1 IJ −1
∴ f(z) =
FG z IJ −
FG 1 IJ =−
2
1−
H2 K −
z H
1−
z K
− 2 1−
H 2 K H
z 1−
z K
∑ FGH 2 IJK ∑ FGH z IJK
∞ n ∞ n
1 z 1 1
=– −
2 n=0
z n=0
This is a series in positive and negative powers of z, so it is an expansion of f(z) in
Laurent’s series within the annulus 1 < | z | < 2
www.cgaspirants.com
www.cgaspirants.com
(c) When |z| > 2
f(z) =
1 1 1FG 2 IJ −1
1 FG
1 IJ −1
∴
FG 2
JKI
−
FG 1
JKI
=
z H
1−
z K −
z
1−
H
z K
H
z 1−
z
z 1−
H z
∑ FGH z IJK ∑ FGH z IJK

∞ n ∞ n
1 2 1 1
= −
z n=0 z n=0
This is Laurent’s series within the annulus 2 < | z | < R, where R is large.
(d) When 0 < |z – 1| < 1
1 1 1 1
f(z) = − =− − = – (z – 1)–1 – [1 – (z – 1)]–1
m
( z − 1) − 1 z − 1 1 − ( z − 1) z − 1
∞
1
=– − ∑
( z − 1) n .
o
z − 1 n=0
.c
This is also Laurent’s series within the annulus 0 < | z – 1 | < 1.
∞
ts ∑ (n + 1)(z + 1) n
Example 2. Show that when | z + 1 | < 1, z–2 =1+ .
n=1
n
1 1 1
Sol. f(z) = z–2 = 2
= 2
= = [1 – (z + 1)]–2
z [( z + 1) − 1] [1 − ( z + 1)]2
ra
= 1 + 2(z + 1) + 3(z + 1)2 + 4(z + 1)3 + ⋅⋅⋅

[By binomial theorem, since | z + 1 | < 1]
pi
∞
=1+ ∑ (n + 1)( z + 1) n
.
as
n=1
π
Example 3. Expand cos z in a Taylor’s series about z = .
4
g
Sol. Here f(z) = cos z, f ′(z) = – sin z, f ″(z) = – cos z, f ″′(z) = sin z, ...
.c
FG π IJ = 1 FG π IJ = − 1 FG π IJ = − 1 FG π IJ = 1
∴ f
H 4K 2
, f′
H 4K , f″
H 4K , f ″′
H 4K , ...
w
2 2 2
Hence cos z = f(z)
FG z − π IJ FG z − π IJ
w
2 3
F πI F πI F πI H 4K F πI H 4K FG π IJ + ...
= f G J + Gz − J f ′ G J +
H 4 K H 4 K H 4 K 2 ! f ″ GH 4 JK + 3 !
w
f ″′
H 4K
1 L F π I 1 FG z − π IJ + 1 FG z − π IJ + ⋅ ⋅ ⋅OP 2 3
= M
2 MN
1 − Gz − J −
H 4 K 2 ! H 4 K 3 ! H 4 K PQ
sin z
Example 4. Expand the function about z = π.
z−π
Sol. Putting z – π = t, we have
sin z sin (π + t) − sin t 1 t3 t5 F I
z−π
=
t
=
t
=–
t
t− +
3! 5!
⋅⋅⋅ GH JK
t2 t4 ( z − π) 2 ( z − π) 4
= −1+ − + ⋅⋅⋅ = − 1 + − + ⋅⋅⋅
3! 5! 3! 5!
www.cgaspirants.com
www.cgaspirants.com
z
Example 5. Expand f(z) = about z = – 2.
(z + 1)(z + 2)
Sol. To expand f(z) about z = – 2, i.e., in powers of z + 2, we put z + 2 = t.
z t−2 2−t 2−t
∴ f(z) = = = = (1 – t)–1
( z + 1)( z + 2) (t − 1) t t(1 − t) t
2−t
= (1 + t + t2 + t3 + ⋅⋅⋅) for 0 < | t | < 1
t
1 2
= (2 + t + t2 + t3 + ⋅⋅⋅) = + 1 + t + t2 + ⋅⋅⋅
t t
2
= + 1 + (z + 2) + (z + 2)2 + ⋅⋅⋅ for 0 < | z + 2 | < 1
z+2
m
which is Laurent’s series.
Example 6. Expand the following function in a Laurent’s series:
o
ez 1
.c
(i) f(z) = about z = 1. (ii) f(z) = for | z – 1 | < 1
(z − 1) 2 z(z − 1) (z − 2)
Sol. (i)
Put
f(z) =
ez
( z − 1) 2
z – 1 = t then
n
z=1+t
ts
LM1 + t + t + t + ⋅ ⋅ ⋅OP = e L 1 + 1 + 1 + t + ⋅ ⋅ ⋅O
ra
1+ t 2 3
e e
∴ f(z) =
t2
=
t2 MN 1 ! 2 ! 3 ! PQ MN t t 2 ! 3 ! PQ 2
LM 1 O
pi
1 1 z−1
=e + + + + ⋅ ⋅ ⋅P .
MN (z − 1) 2
z −1 2! 3! PQ
as
1 1 1 1
(ii) f(z) = = − + |Partial fractions
z ( z − 1) ( z − 2) 2 z z − 1 2 ( z − 2)
g
1 1 1
= − +
.c
2 ( z − 1 + 1) z − 1 2 ( z − 1 − 1)
1 1 1
w
= {1 + (z – 1)}–1 − − {1 – (z – 1)}–1 |∵ |z – 1| < 1

2 z−1 2
w
∞ ∞
1 1 1
=
2 ∑ (− 1) n ( z − 1) n − −
z−1 2 ∑ (z − 1) n
w
n=0 n=0
This is a series in positive and negative powers of (z – 1) hence it is an expansion of f(z)
in a Laurent’s series for |z – 1| < 1.
Example 7. Expand the following function in a Laurent’s series about the point z = 0:
1 − cos z
f(z) = .
z3
Sol. f(z) =
1 − cos z 1
= 3 1− 1−
z2 z4
+
LM
− ⋅⋅⋅
|RS |UVOP
z 3
z 2! 4! MN |T |WPQ
1 F z − z + z – ⋅ ⋅ ⋅I = 1 − 1 z + 1 z
2 4 6
=
z3
GH 2 ! 4 ! 6 ! JK 2 ! z 4 ! 6 ! 3 – ⋅⋅⋅
www.cgaspirants.com
www.cgaspirants.com
1
Example 8. Find the terms in the Laurent’s expansion of for the region
z(e z − 1)
0 < | z | < 2π.
1 1
Sol. f(z) = =
z
z(e − 1) LM
z 1+ z +
z
+
2
z3
+ ... − 1
OP
MN 2! 3! PQ
F z + z + ...I = z F1 + z + z + ...I
2 3 −1 2 −1
= z–1 z + GH 2 ! 3 ! JK GH 2 ! 3 ! JK
–2
LM F z z z z I 1 F z z I 2
m
2 3 4 2
MN1 − GH 2 + 6 + 24 + 120 + ...JK + 4 z GH1 + 3 + 12 + ...JK

2
= z–2
o
1 F z I z (1 + ...) + ...OP
− z G 1 + + ...J +
3 4
.c
3 4
8 H 3 K 16 PQ
LM1 − z + z FG 1 − 1IJ − z FG 1 − 1 + 1 IJ
= z–2 2
N 2 H 4 6 K H 8 6 24 K
n 3
ts
F 1 − 1 + 1 + 1 − 1 IJ + ...OP
+z G
H 16 8 24 36 120 K Q
4
ra
LM1 − z + z − z + ...OP = z – 1 z + 1 − z + ...

2 4
−1
2
pi
= z–2 –2
MN 2 12 720 PQ 2 12 720
as
1
The singularities of are given by z = 0, ez = 1 i.e., z = 0, ± 2πi, ± 4πi, ....... .
z
z(e − 1)
Hence the above expansion is valid for the region 0 < | z | < 2π.
g
Example 9. Using Taylor’s theorem, show that:

.c
(z − 1) 2 (z − 1) 3
log z = (z – 1) – + – ... where | z – 1 | < 1.
w
2 3
Sol. f(z) = log z, f(1) = 0 |∵ a = 1 and log 1 = 0
w
1
Now, f ′(z) = , f ′(1) = 1
z
w
1
f ″(z) = – 2 , f ″(1) = – 1
z
2
f ″′(z) = 3 , f ″′(1) = 2
z
−6
f (iv) (z) = 4 , f (iv) (1) = – 6 and so on.
z
We know that,
( z − a) 2 ( z − a) 3
f(z) = f(a) + (z – a) f ′(a) + f ″ (a) + f ″′(a) + ⋅⋅⋅
2! 3!
( z − 1) 2 ( z − 1) 3
= f(1) + (z – 1) f ′(1) + f ″ (1) + f ″′(1) + ⋅⋅⋅
2! 3!
www.cgaspirants.com
www.cgaspirants.com
( z − 1) 2 ( z − 1) 3
= 0 + (z – 1) (1) + (– 1) + (2) + ⋅⋅⋅
2! 3!
( z − 1) 2 ( z − 1) 3
= (z – 1) – + – ⋅⋅⋅
2 3
Example 10. Find the Taylor’s or Laurent’s series which represent the function
1
2
when
(1 + z )(z + 2)
(i) | z | < 1 (ii) 1 < | z | < 2 (iii) | z | > 2.
Sol. Let f(z) =

1
=
1 RS
1
−
z−2 UV
2
(1 + z )( z + 2) T
5 z + 2 1 + z2 W
m
(i) | z | < 1
1 1 1 FG 2− z IJ −1
o
f(z) = .
5 2
1+ z
2 H +
5
(1 + z2)–1
K
.c
| Binomial expansion of (1 + z)–1 is valid only when | z | < 1
FG z IJ
ts
∞ n ∞
1 2− z
=
10 ∑ (− 1) n
H 2K +
5 ∑ (− 1) n
z2n
n
0 0
This is a series in positive powers of z, so it is an expansion of f(z) in a Taylor’s series
ra
within the circle | z | = 1.
∞ ∞
Remark. If | z | < 1, (1 + z)–1 = ∑ (− 1) n z n ; (1 – z)–1 = ∑z n

.
pi
n=0 n=0
(ii) 1 < | z | < 2
as
1 1 1 FG IJ −1
2− z 1 1 FG IJ −1
f(z) = .
5 2
1+ z
2 H K +
5
. 2 1+ 2
z z H K
g
1
∑
∞
FG z IJ n
2– z
∑
∞
FG 1 IJ n
.c
(– 1) n (– 1) n
=
10 0
H 2K +
5 z2 0
Hz K2
w
This is a series in positive and negative powers of z, so it is an expansion of f(z) in a

Laurent’s series within the annulus 1 < | z | < 2.
w
(iii) | z | > 2
FG IJ –1
FG IJ –1
w
1 1 2 1 1 1
f(z) = .
5 z
1+
z H K –
5
( z – 2) 2 1 + 2
z z H K
1
∑ (− 1) FGH z IJK
∞
2 n
n
FG
1 1 2 IJ ∑ (− 1) FG 1 IJ
∞
n
n
=
5z 0
−
H−
5 z z2 K 0
Hz K 2
This is Laurent’s series within the annulus 2 < | z | < R, where R is large.
Example 11. Find the Taylor’s and Laurent’s series which represent the function
z2 − 1
when (U.K.T.U. 2011)
(z + 2)(z + 3)
(i) | z | < 2 (ii) 2 < | z | < 3 (iii) | z | > 3.
www.cgaspirants.com
www.cgaspirants.com
z2 − 1 3 8
Sol. Let f(z) = =1+ − .
( z + 2)( z + 3) z+2 z+3
(i) | z | < 2
3 FG
z IJ −1
8 FG
z IJ −1
f(z) = 1 +
2
1+
2H K −
3
1+
H
3 K
3
∑
∞
(− 1) n
FG z IJ n
8
∑
∞
(− 1) n
FG z IJ n
=1+
2 0
H 2K −
3 0
H 3K
It is a Taylor’s series within a circle | z | = 2.
(ii) 2 < | z | < 3
m
3 2 8FG z IJ −1
FG IJ −1
f(z) = 1 +
1+ −
H
1+
K H K
o
z z 3 3
| Arranging suitably to make the binomial expansion valid for 2 < | z | < 3
.c
3
∑
∞
(− 1) n
FG 2 IJ n
8
∑
∞
(− 1) n
FG z IJ n
H zK H 3K
ts
=1+ −
z 0
3 0
It is a Laurent’s series within the annulus 2 < | z | < 3.
n
(iii) | z | > 3
ra
3 FG
2 IJ −1
8 FG
3 IJ −1
f(z) = 1 +
z
1+
H
z K −
z
1+
H
z K
pi
3
∑
∞
(− 1) n
FG 2 IJ n
8
∑
∞
(− 1) n
FG 3 IJ n
=1+
z H zK −
z H zK
as
0 0
It is a Laurent’s series within the annulus 3 < | z | < R, where R is large.
1
g
Example 12. Expand in the regions

(z + 1)(z + 3)
.c
(i) | z | < 1 (ii) 1 < | z | < 3

(iii) | z | > 3 (iv) 1 < | z + 1 | < 2.
w
Sol. f(z) =
1
=
1 1
−
1 LM OP
N Q
w
( z + 1)( z + 3) 2 z + 1 z + 3
(i) | z | < 1
w
1 LM
(1 + z) −1 −
1 z FG IJ OP = 1 LM∑ (− 1)
−1 ∞
n
zn −
1
∞
∑ (− 1) n
FG z IJ OP
n
f(z) =
2 MN 3
1+
3 H K PQ 2 MN 0
3 0
H 3 K PQ
It is a Taylor’s series within a circle | z | = 1.
(ii) 1 < | z | < 3
1 1 LM FG
1 IJ −1
1 FG
z IJ −1 OP
MN H K H K
f(z) = . 1+ − 1+
2 z z 3 3 PQ
LM (− 1) F 1I
1 1
∞ n
1
∞
FG z IJ OP
n
=
2 zMN ∑ GH z JK n
−
3 ∑ (− 1) n
H 3 K PQ
0 0
It is a Laurent’s series within the annulus 1 < | z | < 3.
www.cgaspirants.com
www.cgaspirants.com
(iii) | z | > 3
LM FG IJ − 1 FG 1 + 3 IJ OP
1 1 1
−1 −1
MN H K z H z K PQ
f(z) = 1+
2 z z
1 L1 F 1I 1 F 3I O
∞ n ∞ n
= M ∑ (− 1) G J − ∑ (− 1) G J P n n
2 MN z H zK z
0
H z K QP 0
It is a Laurent’s series within the annulus 3 < | z | < R where R is large.

(iv) 1<|z+1|<2
⇒ 1<|u|<2 where z + 1 = u
1 1 LM1 1 1 1 OP LM OP
m
f(z) = − = −
N
2 z+1 z+3 2 u u+2 Q N Q
o
1
.
2 1 1 u FG IJ −1
1
∑
∞
(− 1) n
FG u IJ n
= = = 1+
H K =
H 2K
.c
2 u(u + 2) u(u + 2) 2u 2 2u 0
∑ (− 1) FGH 2 IJK
∞ n
=
1
2( z + 1)
n 0
It is Laurent’s series in the annulus 1 < | z + 1 | < 2.
z+1 n
ts
ra
Example 13. Find the Laurent’s expansion for:
7z − 2
f(z) = (U.K.T.U. 2010)
pi
z − z 2 − 2z
3
in the regions given by:

as
(i) 0 < | z + 1 | < 1 (ii) 1 < | z + 1 | < 3 (iii) | z + 1 | > 3.

Sol. We have
g
7z − 2 1 3 2 1 3 2
f(z) = 3 2
= − + = − +
z − z − 2z z z+1 z−2 ( z + 1) − 1 z + 1 ( z + 1) − 3
.c
(i) 0 < | z + 1 | < 1

w
f(z) = – {1 – (z + 1)}–1 –
3 2 RS FG
z+1 IJ UV −1
−
z+1 3 T H
1−
3 KW
w
3
∑
( z + 1) n
∞
2
− ∑G
F z + 1IJ
∞ n
w
=– −
z + 1 n=0 3 H 3 K
n=0
This is a series in negative and positive powers of (z + 1) hence it is an expansion of f(z)

in Laurent’s series within the annulus 0 < | z + 1 | < 1.
(ii) 1 < | z + 1 | < 3
f(z) =
1 1 FG
IJ −1
3 RS FG IJ UV
2 z+1
−1
z+1
1−
z+1 H
K − −
z+1 3T H KW
1−
3
1
∑
FG 1 IJ ∞ n
3 2
− ∑G
F z + 1IJ ∞ n
=
z+1 H z + 1K
n=0
−
z+1 3 H 3 K n=0
www.cgaspirants.com
www.cgaspirants.com
This is also a series in negative and positive powers of (z + 1) hence it is an expansion of

f(z) in Laurent’s series within the annulus 1 < | z + 1 | < 3.
(iii) | z + 1 | > 3
f(z) =
1
1−
1 FG IJ −1
−
3
+
2
1−
3 IJ
FG −1
z+1 z+1 H K z+1 z+1 z+1 H

K
=
1
∑
FG 1 IJ ∞ n
−
3
+
2
∑
FG 3 IJ ∞ n
z+1 H z + 1K
n=0
z+1 z+1 H z + 1K
n=0
This is a series in negative powers of (z + 1) hence it is an expansion of f(z) in Laurent’s

series within the annulus 3 < | z + 1 | < R where R is large.
m
1
Example 14. (i) Obtain the Taylor’s series expansion of f(z) = 2 about the
z − 4z + 3
o
point z = 4. Find its region of convergence.
.c
1
(ii) Obtain Taylor’s series expansion of f(z) = 2 about the point z = – i. Find the
z +4
region of convergence.
ts (U.P.T.U. 2006)
Sol. (i) If the centre of the circle is at z = 4, then the distances of the singularities z = 1
n
and z = 3 from centre are 3 and 1. Hence if a circle is drawn with centre at z = 4 and radius 1
then within a circle | z – 4 | = 1, the given function f(z) is analytic hence it can be expanded in
ra
Taylor’s series within the circle | z – 4 | = 1 which is therefore the circle of convergence.
pi
Y
as
Cº|z–4|=1
g
O z=1 z=2 z=3 z=4 X

.c
w
w
1 1 1 1 LM1 1 OP1 LM OP
w
f(z) = = − = −
( z − 1)( z − 3) 2 z−3 z−1 N Q
2 z−4+1 z−4+3 N Q
LM1
{1 + ( z − 4)} −1
1R F z − 4 IJ UV
− S1 + G
−1 OP
=
MN
2 3T H 3 KW PQ
1L FG z − 4 IJ OP
∞ ∞ n
f(z) = M ∑ (− 1)
1
⇒
2M
n
( z − 4) n − ∑ (− 1) n
H 3 K PQ
N n=0
3 n=0
www.cgaspirants.com
www.cgaspirants.com
(ii) If the centre of the circle is at z = – i, then

the distances of the singularities z = 2i and z = – 2i
from centre are 3 and 1 respectively.
Hence if a circle is drawn with centre at z = – i and
radius 1 then within a circle |z + i| = 1, the given
function f(z) is analytic hence it can be expanded in
Taylor’s series within the circle |z + i| = 1 which is
therefore the circle of convergence.
1 1
f(z) = 2
=
z +4 ( z − 2i) ( z + 2i)
FG
1 1 1 IJ
m
= −
H
4i z − 2i z + 2i K
1 L OP
o
1 1
= M −
4i N ( z + i) − 3i ( z + i) + i Q
.c
1 L− 1 R F z + i IJ UV − 1 RS1 + FG z + i IJ UV OP
−1 −1
M
=
4i M 3i T
N
S 1− G
H 3i K W i T H i K W
n ts OP
PQ
1 Li FG z + i IJ + i ∑ ( − 1) FG z + i IJ
∞ n ∞ n
M ∑
ra
n
N H 3i K H i K
=
4i M 3 n=0 n=0
PQ
1 L1 F z + i IJ + ∑ ( − 1) FG z + i IJ OP
pi
∞ n ∞ n
= M ∑ G n
N H 3i K
4 M3 n=0
H i K PQ
n=0
as
Example 15. (i) If the function f(z) is analytic and one-valued in | z – a | < R, prove that
for 0 < r < R,
g
1
z 2π
.c
f ′(a) = P( θ ) e − iθ dθ where P(θ) is the real part of (a + reiθ).

πr 0
w
1 FG 1 IJ ∞
H
z t−
K
(ii) Prove that: e 2 t
= ∑ J n (z) t n , | t | > 0
w
n=−∞
z
w
1 π
where Jn (z) = cos (nθ − z sin θ) dθ [G.B.T.U. (C.O.) 2008]
π 0
Sol. (i) ∵ f(z) is regular in | z – a | < R
∴ f(z) is regular in | z – a | = r |∵ r<R
∴ f(z) can be expanded in a Taylor’s series within the circle | z – a | = r. Thus,
∞
f(z) = ∑a
0
m (z − a) m where z – a = reiθ
∞
= ∑a
0
m r m e miθ ...(1)
www.cgaspirants.com
www.cgaspirants.com
∞
⇒ f ( z) = ∑a 0
m r m e − miθ ...(2)
Now, z
C
f ( z)
dz
( z − a) n + 1
= z
0
2π ∞
∑a
0
m r m e − miθ
r
re iθ idθ
n + 1 i( n + 1)θ
e
=
∞
∑a
0
m r m− n i z0
2π
e − i ( m + n ) θ dθ
= 0, for all values of n ...(3)
Particularly, z f ( z)
dz
=0 ...(4)
m
C ( z − a) 2
1
z f ( z) dz 1
z f ( z) + f ( z )
o
We know that f ′(a) = = dz | Using (4)
2πi C ( z − a) 2 2πi C ( z − a) 2
.c
=
1
2πi z 2π
0
f (a + re iθ ) + f (a + re iθ )
2
r e
ts
2 iθ
re iθ idθ
z 2 Re (a + re iθ )
n
1 2π
= dθ |∵ z + z = 2 Re (z)
2π 0 re iθ
ra
=
1
πr z 2π
0
P(θ) e − iθ dθ where P(θ) = Re (a + reiθ).
pi
1 FG
1 IJ
(ii) The function e 2
z t−
tH K is analytic everywhere in the t-plane except at t = 0 and t = ∞
as
i.e., it is analytic in the ring shaped region r ≤ | t | ≤ R where r is small and R is large.
Therefore this function can be expanded in Laurent’s series in the form
g
1 FG1 IJ ∞ ∞
z t−
H K
2 t
∑ an t n + ∑ bn t − n
.c
e =
n=0 n=1
w
z z
1 FG1 IJ 1 FG 1 IJ
where an =
1
e 2
z t−
Ht K dt
and bn =
1
e 2 H
z t−
t K dt
w
n+1 − n+ 1
2πi C t 2πi C t
where C is any circle with centre as origin.
w
Taking C ≡ | t | = 1 so that t = eiθ and dt = ieiθ dθ, we get
an =
1
2πi z 2π
0
e2
z
( eiθ − e − iθ ) ie iθ dθ
e( n + 1) iθ
=
1
2π z
0
2π
e iz sin θ e − niθ dθ =
1
2π z0
2π
e − i( nθ − z sin θ) dθ
=
1
2π z
0
2π
cos (nθ − z sin θ) dθ | Since second part vanishes
⇒ an =
1
π zπ
0
cos (nθ − z sin θ) dθ | Using prop. of definite integrals
www.cgaspirants.com
www.cgaspirants.com
1 1 FG IJ
Clearly, the function e 2
z t−
t H K remains unaltered if t is replaced by –
1
so that
t
bn = (– 1)n an. Therefore,
1 FG 1 IJ ∞ ∞
H
z t−
K
e 2 t
=
n=0
∑ an t n + ∑
n=1
bn t − n
∞ ∞ ∞
= ∑
n=0
an t n + ∑
n=1
(− 1) n an t − n = ∑
n= − ∞
an t n
1 FG 1 IJ ∞
m
H
z t−
K
Here, an is Jn (z) hence, e 2 t
= ∑
n=−∞
J n ( z) t n
o
1 π
where, Jn (z) = cos (nθ − z sin θ) dθ.
.c
π 0
FG IJ ∑ a FGH z IJ
∞
1 n 1
where
Example 16. Prove that cosh z +
an =
1
2π z
0
2π
cosh θ cosh (2 cos θ) dθ.
n H z K
= a0 +
1
ts n +
zn K
(M.T.U. 2013)
ra
1 FG IJ
Sol. The function f(z) = cosh z +
z H
is analytic everywhere in the finite part of the
K
pi
plane except at z = 0 i.e., it is analytic in the annulus r ≤ | z | ≤ R where r is small and R is

large. Hence f(z) can be expanded in Laurent’s series in the annulus r < | z | < R. Thus,
as
FG 1IJ ∑ a z + ∑ b z
∞
n
∞
−n
H
cosh z +
z K
= n n
g
0 1
F 1I FG IJ
.c
1
cosh G z + J
H z K dz
z z H K dz
cosh z +
1 1 z
where a = and bn =
w
n
2πi C z n + 1
2πi C z − n+1
where C is any circle lying in the annulus with origin as centre.
w
FG 1 IJ
z H K dz
cosh z +
w
1 z
an =
2πi C z n+1
=
1
2 πi z 0
2π cosh (2 cos θ) ie iθ dθ
e i(n + 1)θ
| Take C as a circle | z | = 1 on which z = eiθ
=
1
2π z 2π
0
cosh (2 cos θ) e–inθ dθ
=
1
2π z 2π
0
cosh (2 cos θ) cos nθ dθ ∵ z0
2π
cosh (2 cos θ) sin nθ dθ = 0
www.cgaspirants.com
www.cgaspirants.com
bn = a–n
=
1
2π z 0
2π
cosh (2 cos θ) cos (– nθ) dθ
=
1
2π z 0
2π
cosh (2 cos θ) cos nθ dθ = an
FG 1 IJ ∑ a z + ∑ b z
∞
n
∞
−n
∞
∑ an z n +
∞
∑a z −n
Hence, cosh z +
H z
=
K 0
n
1
n =
0 1
n |∵ a n = bn
∞ ∞ ∞
∑ ∑ ∑ a (z
m
= a0 + an z n + an z − n = a0 + n
n
+ z−n ) .
1 1 1
o
Fa I
z
n 2
1 a n e az
Example 17. If C is a closed contour around origin, prove that G J
.c
= dz
H n !K 2 πi C n ! z n+1
F a I 1 e dθ.
Hence deduce ∑ G J =
∞
H n ! K 2π
0
n 2
n
z 2π
0
2a cos θ
ts
ra
Sol. Let f(z) = eaz
∴ f n(z) = an eaz
∴ f n(0) = an
pi
⇒ an =
1
n! z f ( z) dz
as
2πi C z n+ 1
F a I = 1 1 a e dz
z
n 2 n az
GH n ! JK 2πi n ! z
g
⇒ n+1
C
.c
Fa I
z z
∞ n 2 ∞ n az ∞ n az
1 a e 1 a e
∑ GH n ! JK = ∑ 2πi n ! z dz = 2πi ∑ n ! z dz
w
⇒ n+1 n+ 1
C C
0 0 0
w
=
1
e ∑
a
z
1
dz =
1 R| F a I
∞
e S∑ G J
az
|T H z K
n
z az
∞ n
1 dz U|
V|
w
2πi n! z C 2πi 0
n+1
C
0
n! z W
z
FG IJ
z
1
=
1
e az
e ( a / z) dz
=
1
e
H
a z+
z K dz
2πi C z 2 πi C z
=
1
2 πi z 0
2π
e 2 a cos θ
i e iθ dθ
e iθ
where the circle C is taken as
| z|= 1 so that z = e iθ on C
=
1
2π z 0
2π
e 2 a cos θ dθ
∴ dz = i e iθ dθ
www.cgaspirants.com
www.cgaspirants.com
∞
sin θ
Example 18. Prove that for real k, k2 < 1; ∑
n=0
k n sin (n + 1) θ =
1 − 2 k cos θ + k 2
∞
cos θ − k
and ∑
n=0
k n cos (n + 1) θ =
1 − 2k cos θ + k 2
.
Sol.
1 1 k FG IJ −1
1
∑ FGH z IJK
∞
k
n
∑
∞
kn
;|z|>k ...(1)
=
z−k z
1−
z H K =
z n=0
=
n=0
z n+1
Again, put z = e iθ in (1),
∞
1 1
∑
k n e– (n+ 1) iθ
m
= =
z − k e iθ − k n = 0
o
∞
1
⇒ =
cos θ + i sin θ − k n = 0 ∑
k n [cos (n + 1) θ – i sin (n + 1) θ]
.c
∞
(cos θ − k) − i sin θ
⇒
1 − 2 k cos θ + k 2
=
n=0
n ∑
k n [cos (n + 1) θ – i sin (n + 1) θ]
Comparing real and imaginary parts of (2), we get the required results.
ts ...(2)
ra
1 1 7
Example 19. (i) Show that cosec z = + z+ z 3 + ...... ; 0 < | z | < π.
z 3! 360
pi
a
(ii) Find the Taylor’s series expansion of f(z) = about the point z0.
bz + c
as
1
Sol. (i) cosec z = has singular points at z = 0, ± nπ.
sin z
g
We expand the series in 0 < | z | < π.

.c
1 1 1
cosec z = = =
sin z z z 3
5
z 7
z2 z4 z6 LM F
I OP
MN GH
JK PQ
w
z− + − + ... z 1− − + + ......
3! 5! 7! 3! 5! 7!
w
1L F z − z + z + ......I + F z − z + z + ......I + ......OP 2
= M1 + G
2 4 6 2 4 6
zM J G J PQ
N H 3! 5! 7! K H 3! 5! 7! K
w
1L z 2
z 4
z O 1 z + RS 1 − 1 UV z + ......
4
= M1 + − + + ......P = + 4
zN 3 ! 5 ! (3 !) Q z 3 ! T (3 !) 5 ! W
2 2
LM OP
(ii) f(z) =
a
=
a
bz + c b( z − z ) + bz + c
=
1
M a
bz + c M 1 + b( z − z ) P
P
0
MN bz + c PQ
0 0
0
0
aL 1 OP b
= M where bz + c = d, = e
d N 1 + e (z − z ) Q
0
0
d
www.cgaspirants.com
www.cgaspirants.com
∞
a
=
d ∑
n=0
(− 1) n e n ( z − z0 ) n if | e (z – z0) | < 1
a
∞
F b I n
1
=
bz0 + c ∑
n=0
(− 1) n
GH bz + c JK
0
( z − z0 ) n if | z – z0 | <
e
.
4z − 1
Example 20. Find Taylor’s series expansion of about the point z = 0.
z4 − 1
(U.P.T.U. 2007; M.T.U. 2012)
4z − 1 4z − 1
Sol. f(z) = =
m
z − 1 ( z − 1) ( z + 1) ( z 2 + 1)
4
FG − 2z + 1IJ
o
=
3 1FG+
5 IJ +
H 2K
H K
.c
4 z−1 4 ( z + 1) z2 + 1
Expanding about the point z = 0, we get
f(z) = –
3
n
4
5
(1 − z) −1 + (1 + z) −1 + − 2 z +
4
ts
1
2
FG
H
IJ e1 + z j
K
2 −1
F 1I
ra
∞ ∞ ∞
3 5
∑ zn + ∑ (− 1) n z n + G − 2z + J ∑ (− 1) n z 2 n .
=–
4 n=0
4 n=0
H 2K n=0
pi
ASSIGNMENT
as
Expand the following functions as a Taylor’s series (1–3):

g
1. (i) log (1 + z) about z = 0 [U.P.T.U. (C.O.) 2008]

.c
(ii) tan–1 z in powers of z [U.P.T.U. (C.O.) 2009]

(iii) sin–1 z in powers of z (U.P.T.U. 2007)
w
π π
2. (i) sin z about z = (ii) tan–1z about z = (U.P.T.U. 2015)
4 4
w
z
3. about z = 2.
( z + 1)( z + 2)
w
Expand the following functions in Laurent’s series (4–6):

1 1 1
4. , for | z | > 2 5. , for 1 < | z | < 3 6. , for | z | > 2
z−2 z2 − 4 z + 3 z( z − 1)( z − 2)
2 z3 + 1
7. (i) Find Taylor’s expansion of about the point z = 1.
z ( z + 1)
z
( z − 2)
(ii) Define the Laurent series expansion of a function. Expand f (z) = e in a Laurent series
about the point z = 2. (U.P.T.U. 2009)
7-D:\Engineering Mathematics (Bali)\MTU Sem III-IV\MTU 1-7
www.cgaspirants.com
www.cgaspirants.com
( z − 2)( z + 2)
8. Expand f(z) = in the region:
( z + 1)( z + 4)
(a) | z | < 1 (b) 1 < | z | < 4 (c) | z | > 4.
[U.P.T.U. (C.O.) 2008]
1
9. Expand the function f(z) = 2 about (i) z = – 1 (ii) z = 1
z −z−6
z2 − 6 z − 1
10. (i) Find the Laurent’s series expansion of the function f(z) = in the region
( z − 1)( z − 3)( z + 2)
3 < | z + 2 | < 5.
7z − 2
(ii) Find the Laurent’s series expansion of f(z) = in the region 1 < | z + 1| < 3.
z( z + 1) ( z + 2)
m
(G.B.T.U. 2012)
1
11. (i) Obtain the Taylor series expansion of f(z) = about z = 0
z2 + (1 + 2i) z + 2i
o
z
.c
(ii) Expand f (z) = is Laurent series valid for
( z − 1) (2 − z)
(a) | z – 1 | > 1 and (b) 0 < | z – 2 | < 1 (G.B.T.U. 2011, 2013)
(iii) Expand f(z) =
(a) | z – 1 | > 1
z
( z − 1) ( z − 2)
n ts
in Laurent’s series valid for region:
(b) 0 < | z – 2 | < 1 (M.T.U. 2014)

ra
1
12. Find Laurent’s series of f(z) = about its singular points. Determine the region of convergence.
z2 + 1
pi
1
13. Find all possible Taylor’s and Laurent’s series expansions of the function f (z) =
( z + 1) ( z + 2)2
about the point z = 1. Consider the regions
as
(i) | z – 1 | < 2 (ii) 2 < | z – 1 | < 3 (iii) | z – 1 | > 3

1 1
14. The series expansions of the functions and are
g
1− z z−1
FG IJ
.c
1 1 1 1 1
= 1 + z + z2 + ...... and = 1 + + 2 + ......
1− z z−1 z z z H K
w
Adding, we get (1 + z + z2 + ......) +

1 1 1 FG
1 + + 2 + ...... = 0
IJ
z z z H K
w
Is this result true? If not, give the reason.

7 z2 + 9 z − 18
w
15. Expand f(z) = in Laurent series valid for the regions:

z3 − 9 z
(i) 0 < | z | < 3 (ii) | z | > 3 (G.B.T.U. 2013)
z+4
16. If f(z) = , find Laurent’s series expansion in (i) 0 < | z – 1 | < 4 and (ii) | z – 1 | > 4.
( z + 3 ) ( z − 1)2
(M.T.U. 2013)
z
17. Expand f(z) = in Laurent series in 1 < | z | < 2. (G.B.T.U. 2011, 2012)
( z − 1) ( z 2 + 4 )
2
18. Find all Taylor and Laurent series expansion of the following function about z = 0.
− 2z + 3
f(z) = . (U.P.T.U. 2014)
z2 − 3 z + 2
www.cgaspirants.com
www.cgaspirants.com
Answers
z2 z3 z4 z3 z5 z3 3 5
1. (i) z – + − + ...... (ii) z − + − ...... (iii) z + + z + ...
2 3 4 3 5 6 40
1 LM1 + F z − π I − 1 F z − π I − 1 F z − π I + ......OP2 3
2. (i)
2 MN
GH 4 JK 2 ! GH 4 JK 3 ! GH 4 JK PQ
F π I F π I 16 − 64π ( z – π/4) + ...
(ii) tan z = tan G J + G z – J .
–1 −1
2
H 4 K H 4 K π + 16 (π + 16) 2 2 2
FG 1 − 1IJ − FG 1 − 1 IJ (z − 2) + FG 1 − 1 IJ (z – 2) – ......
3.
H 2 3K H 2 3 K 3 2 H2 3 K 5 3
2
m
1
f(z) = ∑ G J
∞
F 2I n
1
5. f(z) = − ∑ G J −
F z I 1 ∑ FG 1IJ ∞ n ∞ n
4.
z H zK 6 H 3 K 2z H z K
o
n=0 n=0 n=0

∞ ∞
.c
n n
1 1 1 1 2
6. f(z) = − +
2z z 2z
n=0 n=0
7. (i) f(z) = 2 z − 2 + ∑
∞
n=0
(− 1) n ( z − 1) n +
n 1
2 ∑
n=0
∞
(− 1) n ts
FG z − 1IJ
H 2 K
n
(ii) f(z) = e
∞
∑
n=0
1 FG
H
2
n! z − 2
IJ
K
n
ra
∑ (− 1) FGH 4 IJK
∞ ∞ n
z
8. (a) f(z) = 1 − ∑
n=0
(− 1) n zn −
n=0
n
pi
1
∞
F 1I
(b) f(z) = 1 − ∑ (− 1) G J − ∑ (− 1) G J n F zI
n ∞
n
n
z H zK H 4K
as
n=0 n=0
(c) f(z) = 1 − ∑ (− 1) FG IJ − ∑ (− 1) FG IJ
∞ n ∞ n
1 1 4 n 4 n
z H zK z H zK
g
n=0 n=0
FG z + 1IJ − 1 ∑ (− 1) (z + 1)
.c
∞ n ∞
1
9. (i) f(z) = –
20 ∑ H 4 K 5
n n
w
n=0 n=0
1 FG z − 1IJ − 1 ∑ (− 1) FG z − 1IJ
∞ n ∞ n
10 ∑ H 2 K
n
(ii) f(z) = –
H 3 K
w
15
n=0 n=0
FG 3 IJ + 1 FG z + 2 IJ + 1 .
w
∞ n ∞ n
1
z + 2 ∑ H z + 2K 5 ∑ H 5 K
10. (i) f(z) =
z+2
n=0 n=0
∞ ∞
9 1 1 8 1
(ii) f(z) = −
z+1 z+1 ∑
n=0
( z + 1) n
∑ (− 1)
–
z+1
n=0
n
( z + 1) n
1 1
LM ∞
FG z IJ − ∑ (− 1) z OP
n ∞
MN ∑ (− 1) n n n
11. (i) f(z) =
1 − 2i 2i
n=0
H 2i K PQ
n=0
∞ ∞
2
(ii) (a) f(z) =
1
z−1
–
2
z−1 ∑
1
( z − 1) n
(b) f(z) =
n=0
∑ (− 1) n
( z − 2) n −
z−2
n=0
www.cgaspirants.com
www.cgaspirants.com
−1 2 F 1 I ∞
2
n ∞
(iii) (a) f(z) = +
z−1 z−1 ∑ GH z − 1 JK n=0
(b) f(z) =
z−2 ∑
− (− 1) ( z − 2) n n
n=0
1
∞
F z − i IJ ;| z – i | < 2 n
2i ( z − i) ∑
12. (i) f(z) = (− 1) G n
n=0
H 2i K
−1 FG z + i IJ ; | z + i | < 2
∞ n
2i ( z + i) ∑ H 2i K
(ii) f(z) =
n=0
1
∞
F z − 1IJ − 1 ∑ (− 1) FG z − 1IJ – 1 (− 1) (n + 1) FG z − 1IJ
n ∞ n ∞ n
2 ∑
(− 1) G
H 3 K 9∑
n n n
13. (i)
H 2 K 3 H 3 K
m
n=0 n=0 n=0
1
∞
FG 2 IJ n
1 FG z − 1IJ – 1 ∑ (− 1) (n + 1) FG z − 1IJ
∞ n ∞ n
∑ (− 1) n ∑ (− 1) n n
o
(ii)
z−1
n =0
H z − 1K −
3 H 3 K 9
n=0
H 3 K n=0
.c
1
∞
FG 2 IJ n
1 F 3 IJ – 1
∞
F 3 I
n ∞ n
(iii) ∑ (− 1) n
H z − 1K −
z−1 ∑
(− 1) G
H z − 1K (z − 1) ∑ (− 1) (n + 1) GH z − 1JK
n n
14.
z−1
n=0 n=0
No. The first series is valid for | z | < 1 and the second series is valid for | z | > 1. There is no
common point where both the series are valid.
n ts 2
n=0
ra
∑ (− 1) FGH 3 IJK ∑ FGH 3 IJK FG 3 IJ F 3I

∞ n ∞ n ∞ n ∞ n
2 1 z 4 z 2 1 4
15. (i) f(z) = +
z 3
n
−
3
(ii) f(z) = +
z z ∑ ( − 1)n
H zK +
z ∑ GH z JK
pi
n=0 n=0 n=0 n=0
∑ (− 1) FGH 4 IJK
∞ n
1 z−1 n 1 5
as
16. (i) f(z) = − +

64 16 ( z − 1) 4( z − 1) 2
n=0
F 4 I
g
∞ n
1 1 5
(ii) f(z) = ∑ (− 1) n GH z − 1 JK − +
.c
16( z − 1) 16( z − 1) 4( z − 1) 2
n=0
LM OP
w
F 1I FG 1 IJ
∑ FGH 2i IJK FG z IJ
∞ n ∞ n ∞ n ∞ n
1 1 1
MMN ∑ GH z JK ∑ 1 z 1
17. f(z) =
10 z
+
z
(− 1) n
H zK +
2i
−
2i ∑ (− 1) n
H 2i K PPQ .
w
n=0 n=0 n=0 n=0
∑ FGH 2 IJK
w
∞ ∞ n
1 z
18. (i) f(z) = ∑
n=0
zn +
2
n=0
;|z|<1
∑ FGH z IJK ∑ FGH 2 IJK

∞ n ∞ n
1 1 1 z
(ii) f(z) = – + ;1<|z|<2
z 2
n=0 n=0

∞ n ∞ n
1 1 1 2
(iii) f(z) = – − ;|z|>2
z z
n=0 n=0
www.cgaspirants.com
www.cgaspirants.com
1.31 ZERO OF AN ANALYTIC FUNCTION [U.P.T.U. (C.O.) 2008]
A zero of an analytic function f(z) is a value of z such that f(z) = 0.

If f(z) is analytic in the neighbourhood of z = a, then by Taylor’s theorem
f(z) = a0 + a1(z – a) + a2(z – a)2 + a3(z – a)3 + ...... + an(z – a)n + ......∞
If a0 = a1 = a2 = ...... = an–1 = 0 but an ≠ 0, then f(z) is said to have a zero of order n at z = a.
The zero is said to be simple if n = 1.
f n (a)
∵ an =
n!
∴ for a zero of order m at z = a,
m
f(a) = f ′(a) = f ″(a) = ...... = f n–1(a) = 0 but f n(a) ≠ 0.
Thus in the neighbourhood of the zero at z = a of order n,
o
f(z) = an(z – a)n + an+1(z – a)n+1 + ......= (z – a)n [an + an+1 (z – a) + ......] = (z – a)n φ(z)
.c
where φ(z) = an + an+1(z – a) + ...... is analytic and non-zero at and in the neighbourhood of
z = a.
1.32 SINGULARITY
n ts
[M.T.U. 2013, U.P.T.U. (C.O.) 2008, 2009]
A singularity of a function f(z) is a point at which the function ceases to be analytic.

ra
1.33 ISOLATED AND NON-ISOLATED SINGULARITY [M.T.U. 2012]

pi
If z = a is a singularity of f(z) and if there is no other singularity within a small circle surrounding
the point z = a, then z = a is said to be an isolated singularity of the function f(z), otherwise it
as
is called non-isolated.
z+1
Example. Consider the function f(z) = .
z( z − 2)
g
It is analytic everywhere except at z = 0 and z = 2. Thus z = 0 and z = 2 are the only

.c
singularities of this function. There are no other singularities of f(z) in the neighbourhood of
z = 0, z = 2. Hence z = 0 and z = 2 are the isolated singularities of this function.
w
Again, consider the function

F πI
w
1
= cot G J
f(z) =
F πI
tan G J
H zK
H zK
w
F πI
It is not analytic at the points where tan G J = 0 = tan nπ i.e., at the points where = nπ
π
H zK z
1
⇒ z= (n = 1, 2, 3, ......)
n
1 1
Thus z = 1, , ,......, z = 0 are the singularities of the function all of which are isolated
2 3
except z = 0 because in the neighbourhood of z = 0, there are infinite number of other singularities
1
z= where n is large. Therefore z = 0 is the non-isolated singularity of the given function.
n
www.cgaspirants.com
www.cgaspirants.com
1.34 TYPES OF SINGULARITY
Let f(z) be analytic within a domain D except at z = a which is an isolated singularity. Draw a
circle C with its centre z = a and radius as small as we wish and another large concentric circle
C of any radius R lying wholly within the domain D. Now in the annulus between these two
circles, f(z) is analytic. If z is any point of the annulus, then by Laurent’s theorem,
∞ ∞
f(z) = ∑
0
an ( z − a) n + ∑ b (z − a)
1
n
−n
where 0 < | z – a | < R.
∞
The second term ∑ b (z − a)
1
n
−n
on the RHS is called the Principal Part of f(z) at the
m
isolated singularity z = a. Now there arise three possibilities :
(i) All bn’s are zero ⇒ no term in P.P. (Removable singularity) (M.T.U. 2012)
o
(ii) Infinite number of terms in P.P. (Essential singularity) (M.T.U. 2012)
(iii) Finite number of terms in P.P. (Pole) [U.P.T.U. (C.O.) 2008]
.c
∞
(i) Removable Singularity. Here f(z) = ∑a n (z − a) n which is analytic for | z – a | < R except
n
at z = a. Let ϕ(z) be the sum function of the power series
0
ts ∞
∑a n (z − a) n . Now ϕ(z) differs from

ra
0
f(z) only at z = a, where there is singularity. To avoid this singularity, we can suitably define
f(z) at z = a, so that we have
RSf ( z )
pi
for 0 <|z − a } < R

φ(z) =
Ta 0 for z = a
as
This type of singularity which can be made to disappear by defining the function suitably
is called removable singularity.
sin ( z − a)
g
Example. The function has removable singularity at z = a because

z−a
.c
sin ( z − a)
=
1
( z − a) −
RS
( z − a) 3 ( z − a) 5
+ − ......
UV
T W
w
z−a z−a 3! 5!
( z − a) 2 ( z − a) 4
w
=1– + − ......
3! 5!
has no terms containing negative powers of z – a. However this singularity can be removed
w
and the function made analytic by defining

sin ( z − a)
= 1 at z = a.
z−a
∞
(ii) Essential Singularity. Here the series ∑ ( z − a)
1
−n
does not terminate.
Example. f(z) = sin

FG 1 IJ has essential singularity at z = a, because
H z − aK
sin
FG 1 IJ = 1 − 1 1 1 1
− ......
+
H z − a K z − a 3 ! (z − a)
5 ! ( z − a) 53
has infinite number of terms in the negative powers of z – a.
www.cgaspirants.com
www.cgaspirants.com
∞
(iii) Pole. Here the series ∑ (z − a)
1
n
consists of finite number of terms. Then z = a is said to
be a pole of order m of the function f(z). When m = 1, the pole is said to be simple.
sin ( z − a)
Example. f(z) = has a pole at z = a because
( z − a) 4
sin ( z − a)
=
1
( z − a) −
LM
( z − a) 3 ( z − a) 5 ( z − a) 7
+ − + ......
OP
( z − a) 4
( z − a) 4
N 3! 5! 7! Q
1 1 1 1 1
= 3
+ − ( z − a) − ( z − a) 3 + ......
m
( z − a) 3 ! ( z − a) 5 ! 7 !
has finite number of terms (here first two terms only) in negative powers of z – a.
o
Thus if z = a is a pole of order m of the function f(z), then
.c
∞
b1 b2 bm
f(z) = ∑a
0
n ( z − a) n + +
z − a ( z − a) 2
+ ...... +
( z − a) m
=
1
( z − a) m
n LMF∞
MNGH ∑ a
0
n
I
JK ts OP
( z − a) n + m + {bm + bm − 1 ( z − a) + ......+ b1 ( z − a) m }
PQ
ra
1
= ϕ(z)
( z − a) m
pi
Clearly, ϕ(z) → bm as z → a. Hence ϕ(z) is analytic in the neighbourhood of the pole z = a.

as
1.35 THEOREMS
(1) The limit point of the zeros of a function f(z) is an isolated essential singularity.
g
Proof. Let z1, z2, z3, ...... be an infinite set of zeros of f(z). Let z0 be their limit point.
.c
(i) If z0 is a point of the set, then z0 will be a zero of f(z) and will have in its neighbour-
w
hood, a cluster of zeros. But zeros are isolated, so, z0 cannot be a zero of f(z) unless f(z) is
identically zero in D.
w
(ii) If f(z) is not identically zero in D, then z0 is not a zero of f(z). But z0 is surrounded by
many zeros. So z0 is a singularity. Also z0 is not a pole since f(z) does not tend to infinity in the
w
neighbourhood of z0. Therefore z0 is an essential singularity. But the singularity is isolated

since in the neighbourhood of z0, f(z) is analytic. Hence z0 is an isolated essential singularity.
(2) The limit point of the poles of a function f(z) is a non-isolated essential singularity.
Proof. Let p1, p2, p3,...... be an infinite set of poles of f(z). Let p0 be their limit point.
(i) If p0 is a point of the set, then p0 will be a pole of f(z) and will have in its neighbour-
hood a cluster of poles. But poles are isolated, so, p0 cannot be a pole of f(z).
(ii) p0 cannot be a zero of f(z) since the function is not analytic (has poles) in the neigh-
bourhood of p0. So, p0 is an essential singularity. This singularity is not isolated, since these
are poles around p0. Hence p0 is a non-isolated essential singularity.
www.cgaspirants.com
www.cgaspirants.com
1.36 DETECTION OF SINGULARITY
(1) Removable Singularity: lim f(z) exists and is finite.

z→ a
z2 – a2
Example. f(z) =
z−a
lim f(z) = 2a
z→ a
So, f(z) has a removable singularity at z = a.
(2) Pole: lim f(z) = ∞.
z→ a
z2 + a2
m
Example. f (z) =
z–a
lim f(z) = ∞.
o
z→ a
So, f(z) has a pole at z = a.
.c
Moreover, the pole is said to be of order n, if there are n terms in the principal part.
e z− a LM1 + (z − a) + (z − a) 2 OP
Example.
( z − a) 2
=
=
1
( z − a)
n
1
2
+
N
1
+
1
ts
2!
+ ......
+ ......
Q
( z − a) 2 ( z − a) 2 !
ra
Since there are only two terms in the negative powers of z – a i.e., there are only 2
(a finite number) terms in the principal part of the function. Hence the function has a pole of
pi
order 2.
(3) Essential singularity: lim f(z) does not exist.
as
z→ a
1
Example. lim e z− a does not exist, so f(z) has an essential singularity at z = a.
g
z→a
.c
EXAMPLES
w
Example 1. Find out the zero and discuss the nature of the singularity of
z−2 1
w
f(z) = 2
sin .
z z−1
Sol. Zeros of f(z) are given by f(z) = 0
w
1
⇒ z – 2 = 0, sin =0
z−1
1
⇒ z = 2, = nπ (n = 0, ± 1, ± 2, ......)
z−1
1
⇒ z = 2, 1 + (n = 0, ± 1, ± 2, ......)
nπ
Clearly, z = 1 is an isolated essential singularity.
Poles of f(z) are given by
z2 = 0
⇒ z=0
Hence z = 0 is a pole of order 2.
www.cgaspirants.com
www.cgaspirants.com
Example 2. Show that the function ez has an isolated essential singularity at z = ∞.

1
1 ρ 1 1 1
Sol. Put z= e = 1+ + + ...... ∞
ρ ρ 2 ! ρ2
1
We have an infinite number of terms in the negative powers of ρ. So the function e ρ has
an isolated essential singularity at ρ = 0. This implies that ez has an isolated essential singu-
larity at z = ∞.
1
Example 3. Discuss singularity of at z = 2πi.
1 − ez
m
1
Sol. f(z) =
1 − ez
o
For poles 1 – ez = 0
.c
⇒ ez = 1 = e2nπi
⇒
Clearly, z = 2πi is a simple pole.
n ts
z = 2nπi (n = 0, ± 1, ± 2, ......)
cot πz
ra
Example 4. Discuss singularity of at z = a and z = ∞ . [U.P.T.U. (C.O.) 2008]

(z – a) 2
pi
cot πz cos πz
Sol. f(z) = 2
=
( z – a) sin πz ( z – a) 2
as
For poles sin πz(z – a)2 = 0

⇒ z = a, πz = nπ (n ∈ I)
g
⇒ z = a, n
.c
Clearly z = ∞ is the limit point of these poles. Hence z = ∞ is a non-isolated essential

singularity. Also z = a, being repeated twice, gives a double pole.
w
z − sin z
Example 5. Discuss the nature of singularity of f(z) = at z = 0.
w
z3
LM z − F z − z + z − z + ......I OP
w
3 5 7
1 1
Sol. f(z) =
z 3 (z – sin z) =
z MN
GH 3 ! 5 ! 7 ! JK PQ
3
1 Fz z z3 5 I 1 − z + z − ......
7 2 4
= G
3
− +
z H 3! 5! 7!
− ......J =
K 3! 5! 7!
Since, there is no term in the principal part of given function hence z = 0 is a removable
singularity.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1 π
1. Discuss singularity of at z = . [U.P.T.U. (C.O.) 2008]
sin z − cos z 4
2. Discuss the nature of singularity of the function f(z) = z cosec z at z = ∞.
3. What is the nature of the singularity at z = ∞ of the function f(z) = cos z – sin z?
1
4. Discuss the singularity of the function f(z) = at z = 0.
1
cos
z
m
5. Discuss the nature of singularity of f(z) = sin at z = 0. [U.P.T.U. (C.O.) 2009]
z
e1/ z
o
6. Find the singularity of the function g(z) = . [U.P.T.U. (C.O.) 2009]
z2
.c
7. Prove that the singularity of cot z at z = ∞ is a non-isolated essential singularity.
ts
8. Find the nature of singularities of the following functions:
1 – ez 1
n
(i) at z = ∞ [U.P.T.U. (C.O.) 2008] (ii) cosec at z = 0.
1+ e z z
ra
1
9. Discuss singularity of f(z) = sin at z = 1.
1− z
pi
Answers
as
1. Simple pole 2. Non-isolated essential 3. Isolated essential

4. Non-isolated essential 5. Isolated essential
g
6. Isolated essential singularity (z = 0)

8. (i) Non-isolated essential, (ii) Non-isolated essential 9. Isolated essential.
.c
1.37 DEFINITION OF THE RESIDUE AT A POLE

w
Let z = a be a pole of order m of a one valued function f(z) and γ any circle of radius r with
w
centre at z = a which does not contain any other singularities except at z = a, then f(z) is
analytic within the annulus r < | z – a | < R hence it can be expanded within the annulus in
w
a Laurent’s series in the form

∞ ∞
f(z) = ∑
n=0
an ( z − a) n + ∑ b ( z − a)
n=1
n
−n
...(1)
where an =
1
2πi z
C
f ( z) dz
( z − a) n + 1
...(2)
and bn =
1
z f ( z)
γ ( z − a) − n + 1
2 πi
dz ...(3)
| z – a | = r being the circle γ.
www.cgaspirants.com
www.cgaspirants.com
Particularly, b1 =
1
2πi z
γ
f ( z) dz
The coefficient b1 is called residue of f(z) at the pole z = a. It is denoted by symbol

Res. (z = a) = b1.
1.38 RESIDUE AT INFINITY
Residue of f(z) at z = ∞ is defined as –

anti-clockwise direction.
1
2πi zC
f ( z) dz where the integration is taken round C in
m
1.39 CAUCHY’S RESIDUE THEOREM OR THE THEOREM OF RESIDUES
[M.T.U. 2013, G.B.T.U. (C.O.) 2011]
o
.c
Let f(z) be one valued and analytic within and on a closed contour C except at a finite number
of poles z1, z2, z3, ......, zn and let R1, R2, R3, ......., Rn be respectively the residues of f(z) at these
poles, then
z C
n
f(z) dz = 2πi (R1 + R2 + R3 + ...... + Rn)
ts
ra
= 2πi (Sum of the residues at the poles within C).
Proof. Let γ1, γ2, γ3, ......, γn be the circles with centres at z1, z2, z3, ......, zn respectively and radii
pi
so small that they lie entirely within the closed curve C and do not overlap. Then f(z) is analytic
within the region enclosed by the curve C and these circles. Hence by Cauchy’s theorem for
as
multi-connected regions, we have
z f ( z) dz = z f ( z) dz + z f ( z) dz + z f ( z) dz + ...... + z f ( z) dz
g
C γ1 γ2 γ3 γn
.c
But by definition of residue,

1
z
w
R1 = f ( z) dz
2πi γ1
z
w
⇒ f ( z) dz = 2πi R1
γ1
z
w
Similarly, f ( z) dz = 2πi R
γ2 2
zγ3
f ( z) dz = 2πi R3
zγn
f ( z) dz = 2πi Rn
Hence, zC
f ( z) dz = 2πi R1 + 2πiR2 + 2πi R3 + ...... + 2πi Rn
= 2πi(R1 + R2 + R3 + ...... + Rn).
www.cgaspirants.com
www.cgaspirants.com
1.40 METHODS OF FINDING OUT RESIDUES
(1) If f(z) has a simple pole (i.e., pole of order 1) at z = a, then
Res {f(z)} = Lt (z – a) f(z).

z→a
Since z = a is a pole of order 1, the Laurent’s series becomes

f(z) = a0 + a1(z – a) + a2(z – a)2 + ...... + b1(z – a)–1.
Multiplying both sides by (z – a), we get
(z – a) f(z) = a0(z – a) + a1(z – a)2 + a2(z – a)3 + ...... + b1
m
∴ Lt (z – a) f(z) = b1 = Res {f(z)}
z→ a
o
(2) If f(z) has a pole of order m at z = a, then
.c
1 d m−1
Res {f(z)} = Lt [(z – a)m f(z)]
(m − 1) ! z → a dz m − 1
Since z = a is a pole of order m, the Laurent’s series becomes
n ts
f(z) = a0 + a1(z – a) + a2(z – a)2 + ...... + b1(z – a)–1 + b2(z – a)–2 + ..... + bm(z – a)–m
Multiplying both sides by (z – a)m, we get
ra
(z – a)m f(z) = a0(z – a)m + a1(z – a)m+1 + a2(z – a)m+2 + ......

+ b1(z – a)m–1 + b2(z – a)m–2 + ...... + bm
pi
Differentiating both sides (m – 1) times w.r.t. z and taking the limit as z → a, we get
as
d m−1
Lt [(z – a)m f(z)] = b1(m – 1) !
z→a dz m − 1
g
1 d m− 1
Lt [(z – a)m f(z)] = b1 = Res {f(z)}.
.c
or
(m − 1) ! z → a dz m− 1
Or
w
LM
d m−1 OP
w
1
Res. {f(z)} = {( z − a) m f ( z)}
N
(m − 1) ! dz m − 1 Q z= a
w
(3) If f(z) is of the form given by

φ( z)
f(z) = ; ψ(a) = 0, φ(a) ≠ 0
ψ ( z)
φ(a)
where z = a is the simple pole of f(z), then Residue of f(z) at z = a is = .
ψ ′ (a)
(4) Residue of f(z) at z = a pole {simple or of order m}
1
= Coefficient of in f(a + t) expanded in powers of t, where t, is sufficiently small.
t
www.cgaspirants.com
www.cgaspirants.com
(5) Residue of f(z) at z = ∞

= lim { − zf ( z)}
z→∞
Or
1
LM
= – coefficient of
z
in the expansion of f(z) for values of z in the OP
N neighbourhood of z = ∞ Q.
EXAMPLES
Example 1. Determine the poles of the following functions and residue at each pole:
z2 1 1 – e 2z
m
(i) 2 (ii) 4 (iii)
(z − 1)(z − 2) z +1 z4
z2
o
Sol. (i) f(z) = .
( z − 1)( z − 2) 2
.c
Poles are given by
(z – 1)(z – 2)2 = 0 ⇒ z = 1, 2.
n
z2 z2
ts
z = 1 is a simple pole while z = 2 is a double pole.
Residue of f(z) at simple pole (z = 1) is
(1) 2
R1 = lim ( z − 1) . = lim = = 1.
ra
2
z→1 ( z − 1)( z − 2) 2 z → 1 ( z − 2) (1 − 2) 2
Residue of f(z) at double pole (z = 2) is
LM R UVOP
pi
1 d z2
MN ST
R2 = ( z − 2) 2 .
(2 − 1) ! dz ( z − 1)( z − 2) 2 WPQ z=2
as
L d F z I OP = L (z − 1) . 2z − z
=M
2 2 OP LM z − 2z OP
2
MN dz GH z − 1JK QP MN (z − 1) =0
=
2
Q N (z − 1) Q
2
g
z=2 z=2 z=2

1
.c
(ii) f(z) = 4
z +1
w

z4 + 1 = 0
w
⇒ z = (– 1)1/4 = {e(2n+1)πi}1/4
∴ Poles are, z = e(2n+1)πi/4 where, n = 0, 1, 2, 3, ......
w
These are all of order 1 since the four factors occur linearly in z4 + 1.
Since the roots repeat themselves, we can write them more conveniently as
e(2n+1)πi/4 where, n = – 2, – 1, 0, 1
i.e., emπi/4 where, m = ± 1, ± 3. Let denote it by zm.
Residue at (z = zm) is
1 1
R = lim ( z − zm ) . 4 = lim | By L’ Hospital’s Rule
z → zm z +1 z → zm 4 z3
1 zm zm 1
= 3
= 4
=– = − emπi/4 h where, m = ± 1, ± 3.
4 zm 4 zm 4 4
www.cgaspirants.com
www.cgaspirants.com
1 − e2 z
(iii) Pole of is evidently z = 0. But this is not of the fourth order since
z4
1 − e2 z 1 LM R 4 z 2 8 z 3 16 z 4 UVOP
MN ST
= 1 − 1 + 2 z + + + + ......
z4 z4 2 6 24 WPQ
FG 2 + 2z + 4 z 2 2 3 IJ
=–
H 3
+
3
z + ......
K ...(1)
z3
Therefore the pole is of order 3.
Residue at this pole is
m
R = lim
RS (1 − e ) z UV
1 d2 2z 3
z→0
T z W
2 ! dz 2 4
o
1 L d R| 1 F I U|VOP
.c
2 2 3
= lim M S
2 ! M dz |T z HG 1 − 1 − 2z −
4z
−
8z
− ......JK |WPQ
z→0
N 2
2! 3!
1Ld F
= lim M
z→02 N dz H
G
2
n
2
4
3
IO
− 2 − 2 z − z − ......J P
KQ
2
ts
ra
1L 8 2 O 4
= lim M− − . 6 z − ......P = −
z→02N 3 3 Q 3
pi
Example 2. Find the residue at z = 0 of the following functions:

as
1 + ez 1
(i) (ii) z cos .
sin z + z cos z z
g
Sol. (i) z = 0 is a pole of order 1.

.c
z(1 + e z ) 1 + ez 1+ 1
Residue = lim = lim = = 1.
FG IJ
w
z→0 sin z + z cos z z → 0 sin z 1+ 1

z
+ cos z
H K
w
(ii) Expanding the function in powers of z, we have
LM OP
w
1 1 1 1 1
z cos = z 1− 2 + − ...... = z – + – ......
z N2z 4 ! z4 2 z 24 z 3 Q
This is the Laurent ’s expansion about z = 0.
1 1 1 1
The coefficient ofin it is – . So the residue of z cos at z = 0 is – .
z 2 z 2
Example 3. (a) Give an example of a function having residue at infinity yet analytic
there.
z3
(b) Find the residue of f(z) = at z = ∞.
z2 − 1
www.cgaspirants.com
www.cgaspirants.com
z2
Sol. (a) f(z) =
( z − α)( z − β)( z − γ )
Residue of f(z) at z = ∞ is
= lim − z .
RS z2 UV
z→∞
T ( z − α)( z − β)( z − γ ) W
−1
= lim
z→∞ FG 1 − α IJ FG β IJ FG 1 − γ IJ = –1
H zK H 1−
z K H zK
m
1
1 FG IJ λ2 λ
o
Now, f
λ
=
H K
(1 − αλ) (1 − βα) (1 − γλ)
. .
=
(1 − αλ)(1 − βλ)(1 − γλ)
.c
λ λ λ
FG 1IJ = 0 (≠ ∞)
At λ = 0, f
H λK
n ts
FG 1IJ is analytic at λ = 0
ra
f
∴
H λK
⇒ f(z) is analytic at z = ∞.
pi
F − z . z I which does not exist

3
(b) Required residue = Lt GH z − 1JK
as
z→∞ 2
F 1I F 1 + 1 + ......IJ = z + 1 + 1
g
3 −1
z
= z G1 − J = z G1 +
Hence, f(z) =
F 1I H K H z z K z z + ......
.c
2 2 4 3
z G1 − J 2 z
H zK 2
w
F 1I
Required residue = – G coefficient of J = – 1.
H zK
w
z ez
w
Example 4. Evaluate dz , where C is the circle | z – 1 |= 3.

C (z + 1) 2
ez
Sol. Here f(z) = has only one singular point z = – 1 which is a pole of order 2 and
( z + 1) 2
it lies inside the circle | z – 1 | = 3.
d d z
Residue of f(z) at z = – 1 is Lt [(z + 1)2 f(z)] = Lt ( e ) = Lt ez = e–1
z→−1 dz z → − 1 dz z→−1
∴ By Residue theorem, we have

zC
ez
( z + 1) 2
dz = 2πi(e −1 ) =
2πi
e
.
www.cgaspirants.com
www.cgaspirants.com
z2
Example 5. Determine the poles of the function f(z) = and the residue at
(z − 1) 2 (z + 2)
each pole. Hence evaluate

zC
z2
(z − 1) 2 (z + 2)
dz where C ≡ |z| = 3. (U.P.T.U. 2015)
Sol. The function f(z) has a pole of order 2 at z = 1 and a simple pole at z = – 2.
Residue of f(z) at z = 1 is
d d z2 F I
R1 = Lt
z→1 dz
[(z – 1)2 f(z)] = Lt
z → 1 dz z + 2
GH JK
m
( z + 2) . 2 z − z 2 . 1 z2 + 4 z 5
or R1 = Lt 2
= Lt =
z→1 ( z + 2) z → 1 ( z + 2) 2 9
o
Residue of f(z) at z = – 2 is
.c
z2 4
R2 = Lt [(z + 2) f(z)] = Lt = .
ts
2
z → −2 z → −2 ( z − 1) 9
Since both the poles lie inside the given curve C ≡ | z | = 3,
n
z z2 LM 5 + 4 OP = 2πi.
ra
∴ dz = 2πi ( R1 + R 2 ) = 2πi
C 2
( z − 1) ( z + 2) N9 9Q
| By Cauchy’s Residue theorem
pi
Example 6. Determine the poles of the following function and residues at each pole:
z−1
z
as
f(z) = and hence evaluate f ( z) dz , where C is the circle | z – i | = 2.

(z + 1) 2 (z − 2) C
(U.K.T.U. 2011)
g
Sol. Poles of f(z) are given by

.c
(z + 1)2 (z – 2) = 0 ⇒ z = – 1 (double pole), 2(simple pole)

Residue of f(z) at z = – 1 is
w
1 LM R
d z−1 UVOP Y
MN ST
2
( z + 1) 2 .
w
R1 = |=
(2 − 1) ! dz ( z + 1) 2 ( z − 2) WPQ z =−1
(0, 3)
C≡
| z–
i
w
L d FG z − 1IJ OP
=M =
LM − 1 OP =
−1
N dz H z − 2 K Q z =−1 N (z − 2) Q
2
z =−1
9
2
(0, 1) z = i
Residue of f(z) at z = 2 is 1
– 1 O (0, 0)
z−1 (– 3, 0) 2 X
R2 = Lt ( z − 2) 2
z→2 ( z + 1) ( z − 2) (0, – 1)
z−1 1
= Lt 2
=
z→2 ( z + 1) 9
The given curve C ≡ | z – i | = 2 is a circle whose centre is at z = i [i.e., at (0, 1)] and
radius is 2. Clearly, only the pole z = – 1 lies inside the curve C.
www.cgaspirants.com
www.cgaspirants.com
Hence, by Cauchy’s residue theorem
z C
f ( z) dz = 2πi (R1) = 2πi
FG − 1IJ = – 2πi .
H9K 9
z 2 − 2z
(z + 1) 2 (z 2 + 4)
dz, where C is the circle | z | = 10.
(U.P.T.U. 2009)
Sol. Singularities are given by
(z + 1)2 (z2 + 4) = 0 ⇒ z = – 1 (double pole), ± 2i (simple poles)
All the poles lie inside the given circle c ≡ | z | = 10.
m
∴ Residue (at z = – 1) is
1 dLM R z2 − 2z UVOP
o
MN ST
R1 = ( z + 1) 2
2 − 1 ! dz ( z + 1) 2 ( z 2 + 4) WPQ
.c
z=−1
LM d F z − 2 z I OP
2
L 2z + 8z − 8 OP
=M
2
14
Residue (at z = 2i) is

=
MN dz GH z + 4 JK PQ
n 2
z=−1 ts
N (z + 4) Q 2 2
z=−1
=−
25
ra
z2 − 2 z
R2 = lim ( z − 2i) 2
z → 2i ( z + 1) ( z − 2i) ( z + 2i)
pi
− 4 − 4i 1+ i 7+i
= 2
= =
(2i + 1) (4i) 3i + 4 25
as
Similarly, Residue (at z = – 2i) is

7−i
g
R3 =
25
.c
By Cauchy’s Residue theorem,
z
w
z2 − 2 z
dz = 2πi (R1 + R2 + R3) = 2πi − + +
LM
14 7 + i 7 − i
=0
OP
C ( z + 1) 2 ( z 2 + 4) 25 25 N
25 Q
w
Example 8. Evaluate z 12z − 7

dz , where C is the circle
w
C (z − 1) 2 (2z + 3)
(i) | z | = 2 (G.B.T.U. 2011) (ii) | z + i | = 3.
12 z − 7
Sol. f(z) = .
( z − 1) 2 (2 z + 3)
Poles are given by
3
z = 1 (double pole) and z = – (simple pole)
2
Residue at (z = 1) is
1 d LM R 12 z − 7 UVOP
MN ST
R1 = ( z − 1) 2 .
(2 − 1) ! dz ( z − 1) 2 (2 z + 3) WPQ z= 1
www.cgaspirants.com
www.cgaspirants.com
LM d FG 12z − 7 IJ OP LM (2z + 3) . 12 − (12z − 7) . 2 OP

=
N dz H 2z + 3 K Q z= 1
=
N (2 z + 3)2
Q z=1
60 − 10
= = 2.
25
FG 3 IJ
Residue at simple pole z = −
H 2
is
K
FG z + 3 IJ . 12 z − 7
R2 = lim
z → − 3/ 2 H 2 K ( z − 1) (2 z + 3)
2
1 (12 z − 7)
= lim . = – 2.
z → − 3/ 2 2 ( z − 1) 2
m
3
(i) The contour | z | = 2 encloses both the poles 1 and – .
2
o
∴ The given integral = 2πi (R1 + R2) = 2πi (2 – 2) = 0.
.c
(ii) The contour | z + i | = 3 is a circle of radius 3 and centre at z = – i. The distances
3 13
of the centre from z = 1 and – are respectively 2 and
second is > 3 .
2
n 4
∴ The second contour includes only the first singularity z = 1.

ts
. The first of these is < 3 and the
ra
Hence, the given integral = 2πi (R1) = 2πi (2) = 4πi.
Example 9. Evaluate z z−3

pi
C z 2 + 2z + 5
(i) | z | = 1 (ii) | z + 1 – i | = 2 (iii) | z + 1 + i | = 2.

as
(G.B.T.U. 2013)
z−3
Sol. The poles of f(z) = are given by
g
2
z + 2z + 5
.c
z2 + 2z + 5 = 0 ⇒ z = – 1 ± 2i
(i) Both the poles lie outside the circle | z | = 1.
w
∴ By Cauchy’s integral theorem, we have z z−3

dz = 0
w
2
C z + 2z + 5
(ii) Only the pole z = – 1 + 2i lies inside the circle | z + 1 – i | = 2
w
Residue of f(z) at z = – 1 + 2i is
( z − α)( z − 3) 0
Lt (z + 1 – 2i) f(z) = Lt , where α = – 1 – 2i Form
z → − 1 + 2i 2
z → α z + 2z + 5 0
( z − α) + ( z − 3)
= Lt | By L’ Hospital’s Rule
z→α 2z + 2
α−3 − 1 + 2i − 3 i − 2
= = =
2α + 2 − 2 + 4i + 2 2i
∴ By Cauchy’s residue theorem, 2

z−3
C z + 2z + 5 z dz = 2πi
i−2
2i
= π(i – 2).
FG
H
IJ
K
www.cgaspirants.com
www.cgaspirants.com
(iii) Only the pole z = – 1 – 2i lies inside the circle | z + 1 + i | = 2.

Residue of f(z) at z = – 1 – 2i is
( z − β)( z − 3) 0
Lt (z + 1 + 2i) f(z) = Lt , where β = – 1 – 2i Form
z → − 1 − 2i
2
z → β z + 2z + 5 0
( z − β) + ( z − 3)
= Lt | By L’ Hospital’s Rule
z→β 2z + 2
β−3 − 4 − 2i i + 2
= = =
2β + 2 − 4i 2i
∴ By Cauchy’s residue theorem, z 2

z−3
dz = 2πi
i+2 FG
= π(i + 2)
H
IJ
K
m
C z + 2z + 5 2i
z3
Example 10. Find the residue of f(z) = at its pole and hence evaluate
o
4
(z − 1) (z − 2)(z − 3)
z
.c
f ( z) dz , where C is the circle | z | = 5/2.
C
Sol. Poles of f(z) are given by (z – 1)4 (z – 2)(z – 3) = 0 ⇒ z = 1, 2, 3
z = 1 is a pole of order 4 while z = 2 and z = 3 are simple poles.
n ts
ra
z3 z3 8
R1 = Lt ( z − 2) = Lt = =–8
z→2 ( z − 1) 4 ( z − 2)( z − 3) z → 2 ( z − 1) 4 ( z − 3) (− 1)
pi
z3 z3 27
R2 = Lt ( z − 3) . = Lt
as
4 4
=
z→3 ( z − 1) ( z − 2)( z − 3) z → 3 ( z − 1) ( z − 2) 16
g
1 LM R
d3 z3 UVOP
S
.c
R3 = ( z − 1) 4 .
MN T
(4 − 1) ! dz 3 4
( z − 1) ( z − 2)( z − 3)WPQ z=1
w
1Ld R
= M
3
S z 3 UVOP = 1 LM d RSz + 5 + 19 z − 30 UVOP
3
6 MN dz T ( z − 2)( z − 3) WPQ 6 MN dz T z − 5 z + 6 WPQ

w
3 3 2
z=1 z=1
1Ld R 8 UO 1Ld R UVOP

w
3 2
27
= M
6 N dz T
S
3
z+5+ −
z − 3 z − 2 WQ
V P = M
6 MN dz T
S1 − (z −273) + ( z −82)
2 2 2
WPQ
z=1 z=1
1 L d R 54 16 UO
= M S VP = 61 LM (z− −162 48 O
N 3) (z − 2) PQ
− +
6 MN dz T ( z − 3) 3
( z − 2) WPQ 3 4 4
z=1 z=1
1 L − 162 O 27 = 101
= M + 48P = 8 −
6 N 16 Q 16 16
The given curve C ≡ | z | = 5/2 is a circle with centre at (0, 0) and radius 5/2.
Clearly, only the poles z = 1 and z = 2 lie inside this circle.
www.cgaspirants.com
www.cgaspirants.com
Hence, By Cauchy’s Residue theorem,
zC
f ( z) dz = 2πi (R + R ) = 2πi 101 − 8 = 2πi − 27 = –
3 1
16 16
FG
27πi
H
8
.
IJ
K
FG
H
IJ
K
Example 11. (i) Find the value of z
C
ze 1/z dz around the unit circle.
(ii) Using Residue theorem, evaluate

1
2 πi z
C
e zt
z 2 (z 2 + 2z + 2)
|z| = 3. (U.P.T.U. 2009)

1/z 1/z
Sol. (i) The only singularity of ze is at the origin. Expanding e , we have
m
LM
ze1/z = z 1 +
1 1 1
+ 2 + 3 + ...... = z + 1 +
1 1 OP
+ 2 + ......
N Q
o
z 2z 6z 2z 6z
.c
1 1
Residue at origin = coefficient of = .
z 2
Hence, the required integral = 2πi

n FG 1IJ = πi.
H 2K
ts
ra
(ii) Singularities are given by

z2 (z2 + 2z + 2) = 0 ⇒ z = 0, – 1 ± i
pi
z = 0 is a pole of order 2. z = – 1 ± i are simple poles. All these poles lie inside the circle
|z| = 3.
Residue (at z = 0) is
as
1 d LM |RS e zt |UVOP LM d |R e zt
|UVOP
R1 = z2 .
|T |WPQ = S
MN dz |T 2πi (z + 2z + 2) |WPQ
g
(2 − 1) ! dz MN
2π i z 2 ( z 2 + 2 z + 2)
z=0
2
z=0
.c
LM
1 ( z 2 + 2 z + 2) t e zt − e zt (2 z + 2) O 1 F t − 1I
G J
w
=
2πi MN ( z 2 + 2 z + 2) 2
PPQ =
2πi H 2 K
z=0
w
Let – 1 + i = α and – 1 – i = β then

w
Residue at (z = α = – 1 + i) is
R2 = Lt ( z − α) .
1 e zt
=
1 e αt
=
1 1 (− 1 + i) t
e
LM OP
z→α 2
2πi z ( z − α) ( z − β) 2
2πi α (α − β) 2πi 4 N Q
Residue at (z = β = – 1 – i) is
R3 =
LM
1 1 (− 1 − i) t
e
OP
2πi 4 N Q
www.cgaspirants.com
www.cgaspirants.com
By Residue theorem,
1
2πi z
C 2 2
e zt
z ( z + 2 z + 2)
dz = 2πi
LM 1 FG t − 1 IJ +
N 2πi H 2 K 2πi
R|S
1 e ( − 1 + i) t + ( e ( − 1 − i) t
T| 4
U|VOP
W|PQ
t − 1 e−t
= + cos t .
2 2
1
Example 12. Obtain Laurent’s expansion for the function f(z) = at the isolated
z 2 sinh z
singularity and hence evaluate zC z 2 sinh z

1
dz , where C is the circle | z – 1 | = 2.
m
1 2
Sol. Here, f(z) = =
z 2 sinh z z 2 (e z − e − z )
o
.c
2
=
LMF 1 + z + z + z + ......I − F 1 − z + z − z + ......I OP
2 3 2 3
MNGH JK GH JK PQ
z2
=
2! 3!
n
2
2! 3!
ts
F I
ra
3
2z 2 z5
z2 2 z +GH 3!
+
5!
+ ...... JK Y
| =2
–1
pi
|z
1 C≡
=
F z + z + ......I
2 4
GH 3 ! 5 ! JK
as
z3 1 +
O X
z=1
LM1 + F z + z I + ......OP
g
−1
2 4
MN GH 6 120 JK PQ
= z–3
.c
F 1 − z − z + z + ......I
w
2 4 4
= z–3 GH 6 120 36 JK
w
1 1 7
w
= − + z + ......
z3 6 z 360
Only pole z = 0 of order two lies inside the circle C ≡ | z – 1 | = 2.
1 1
Residue of f(z) at (z = 0) is = coeff. of in the Laurent’s expansion of f(z) = – .
z 6
zC 2
z sinh z
dz
= 2 πi −
FG 1 IJ = – πi .
H 6K 3
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
Determine the poles of the following functions and the residue at each pole:
2z + 1 z+1 ez
1. 2 2. 3. .
z −z−2 z2 ( z − 2) z2 + π 2
Evaluate the following integrals using Cauchy’s residue theorem:
4.
z LMMN
C
cos πz2 + sin πz2
( z + 1) ( z + 2)
OP
dz ; C ≡ | z | = 3
PQ
5.
z LMMN 3 z2 + z + 1 OP dz , where C is the circle | z | = 2.
m
C ( z2 − 1) ( z + 3) PQ
z z2 + 2 z − 2
o
6. dz , where C is a closed curve containing the point z = 4 in its interior.
z−4
z
C
.c
1 − 2z
7. dz , where C is the circle | z | = 1.5.
z( z − 1)( z − 2)
z
C
8.
C
z
( z − 1)( z − 2) 2
dz , where C is the circle | z – 2 | = 21 .
n ts
9.
z sin πz2 + cos πz2
dz , where C is the circle | z | = 3.
ra
C ( z − 1)2 ( z − 2)
10.
z 5z − 2
dz ; C ≡ | z | = 2 11.
z ez − 1
dz ; C ≡ | z | = 1/2
pi
C z( z − 1) C z( z − i) 2 ( z − 1)
12.
z FGH z cos z
dz, where C ≡ | z – 1 | = 1 13.
z dz
, where C ≡ | z – i | = 2
as
C π IJ 3
C ( z2 + 4)2
z−
2 K
z
g
3 z2 + 2
14. dz , where C ≡ | z – 2 | = 2
.c
C ( z − 1)( z2 + 9)
z 2z + 1
z dz
w
15. dz , where C ≡ | z | = 1 16. dz , where C ≡| z | = 1.5

C (2 z − 1)2 C ( z2 + 1)( z2 − 4)
z
w
4 z2 − 4 z + 1
17. (i) dz , where C ≡ | z | = 1.
C ( z − 2)(4 + z2 )
w
(ii) z C
24 z − 7
( z − 1) 2 ( 2 z + 3 )
dz, where C is the circle of radius 2 with centre at the origin.
(M.T.U. 2012)
18.
zC
z2 + 4
2
z( z + 2 z + 2)
dz, where C is
(i) | z | = 1 (ii) | z + 1 – i | = 1
(iii) | z + 1 + i | = 1 (iv) | z – 1 | = 5
19.
zC
e −z
z2
dz ; C ≡ | z | = 1 20.
z C
z2 e1/ z dz ; C ≡ | z | = 1
www.cgaspirants.com
www.cgaspirants.com
21.
zC
2
1
z sin z
dz where C is the triangle with vertices (0, 1), (2 – 2) and (7, 1). (G.B.T.U. 2012)
z
22. Determine the poles and residues at each pole of the function f(z) = and hence evaluate
z2 − 3 z + 2
zC
f ( z ) dz where C is the circle | z – 2| =
1
2
. (G.B.T.U. 2011)
Answers
1 5 3 3 i
1. z = – 1, 2 ; , 2. z = 0, 2 ; – , 3. z = ± πi ; ±
3 3 4 4 2π
πi
4. – 4πi 5. − 6. 44πi
m
4
7. 3πi 8. – 2πi 9. 4πi(π + 1)
o
10. 10πi 11. 0 12. – 2πi
13. π/16 14. πi
.c
15. πi 16. 0 17. (i) 0 (ii) 0
18. (i) 4πi (ii) – π (3 + i) (iii) π (3 – i) (iv) 2πi
19.
22.
– 2πi
z = 1, 2 ; – 1, 2 ; 4πi.
n 20.
πi
3
. ts 21.
2i (− 1) n
n2 π
ra
1.41 CONTOUR INTEGRATION

pi
We take a closed curve C, find the poles of f(z) within C and calculate residue at these poles.
Then by Cauchy’s residue theorem
z
as
f ( z) dz = 2πi [sum of the residues of f(z) at the poles within C]

C
g
The curve is called a contour.

The process of integration along a contour is called contour integration.
.c
1.42 APPLICATION OF RESIDUE THEOREM TO EVALUATE REAL INTEGRALS

w
The residue theorem provides a simple and elegant method for evaluating many important
w
definite integrals of real variables. Some of these are illustrated below.
z
w
2π
1.42.1. Integrals of the Type F(cos θ, sin θ) dθ , where F(cos θ, sin θ) is a Rational
0
Function of cos θ and sin θ.
Such integrals can be reduced to complex line integrals by the substitution z = eiθ, so
that
dz
dz = ieiθ dθ, i.e., dθ = .
iz
e iθ + e − iθ 1 FG IJ
1
Also, cos θ =
2
=
2 H K
z+
z
e iθ – e − iθ 1 F 1I
G z− J.
2i H zK
sin θ = =
2i
www.cgaspirants.com
www.cgaspirants.com
As θ varies from 0 to 2π, z moves once round the unit circle in the anti-clockwise direction.
∴ z
0
2π
F(cos θ, sin θ) dθ = zC
F
Fz+ z
GH 2
−1
,
z − z −1 dz
2i iz
I
JK
where C is the unit circle | z | = 1.
The integral on the right can be evaluated by using the residue theorem.
EXAMPLES
Example 1. Using contour integration, evaluate z 2π dθ

where a > |b|
m
0 a + b cos θ
Hence or otherwise evaluate (U.K.T.U. 2010)
o
(i) z 2π dθ
(ii) z π dθ
; a > |b|
.c
0 2 − cos θ 0 a + b cos θ
Sol. Consider the integration round a unit circle C ≡ |z| = 1 so that z = eiθ
∴ dz = ieiθ dθ = iz dθ
n
⇒ dθ = ts dz
iz
FG IJ
ra
1 iθ 1 1
Also, cos θ = (e + e − iθ ) = z+
2 2 z H K
pi
Then the given integral reduces to
z LM 1 FG dz IJ = z 2z dz FG IJ
H K
as
I=
bF 1I O H iz K
2
bz + 2 az + b iz
a + Gz + JP
C C
N 2H zK Q
g
2
z dz 2
z dz
.c
= =
ib C 2 2a ib C ( z − α) ( z − β)
z + z+1
b
w
a a 2 − b2 a a 2 − b2
where, α= −
+ and β = − −
w
b b b b
Poles are given by (z – α) (z – β) = 0 ⇒ z = α, β
w
Both are simple poles.

Since a > |b| ∴ |β| > 1
Since αβ = 1
∴ |αβ| = 1
|α| |β| = 1
⇒ |α| < 1 |∵ |β| > 1
Hence z = α is the only pole which lies inside the circle C ≡ |z| = 1.
Residue of f (z) at (z = α) is
2
R = Lt ( z − α) f ( z) = Lt ( z − α) .
z→α z→α ib ( z − α) ( z − β)
www.cgaspirants.com
www.cgaspirants.com
2 2 (b) 1
= = =
ib (α − β) ib (2 a 2 − b2 ) i a − b2
2

F I
I = 2πi(R) = 2πi GG 1
JJ
Hi a 2 − b2 K
z
0
2π dθ
a + b cos θ
=
2π
a − b2
2
...(1)
(i) Putting a = 2 and b = – 1 in (1), we get
m
z2π dθ
=
2π
= 2π
o
0 2 − cos θ 2−1
.c
(ii) From (1),
2 zπ
0
dθ
a + b cos θ
=
2π
a − b2
2
ts | Using prop. of definite integrals
z
n
π dθ π
⇒ = .
a + b cos θ
ra
0
a − b2
2
Example 2. Evaluate by contour integration:
z
pi
2π dθ
, where a > |b| [U.P.T.U. (C.O.) 2010; G.B.T.U. 2012]
0 a + b sin θ
as
Hence or otherwise evaluate z0

2π dθ
1 − 2a sin θ + a 2
, 0 < a < 1.
g
Sol. Consider the integration round a unit circle C ≡ |z| = 1

.c
dz
so that z = eiθ ∴ dθ = .
w
iz
1 iθ 1 1 FG IJ
w
Also, sin θ = (e − e − iθ ) =
2i 2i
z−
z H K
w
Then the given integral reduces to
I=
z LM
C b F
1 FG dz IJ =
G z − 1z IJK OPQ H iz K
zC bz 2
2iz FG dz IJ
+ 2iaz − b H iz K
2i H
a+
N
=
2
b z C
z2 +
dz
2ia
b
z−1
Poles are given by

2ia
z2 + z−1=0
b
www.cgaspirants.com
www.cgaspirants.com
− 2ia − 4a2
± +4
b b2 − ia b2 − a 2
⇒ z= = ±
2 b b
2 2
− ia i a − b
= ± = α, β (simple poles)
b b
2 2 2 2
− ia i a − b − ia i a − b
where, α= + and β= −
b b b b
Clearly, |β| > 1
But αβ = – 1
m
∴ |α β| = 1 ⇒ |α| |β| = 1 ⇒ |α| < 1
Hence z = α is the only pole which lies inside circle C ≡ |z| = 1.
o
Residue of f (z) at (z = α) is
2 2
.c
R = Lt ( z − α) . =
z→α b ( z − α) ( z − β) b (α − β)
=
b
n F 2i
GG
2
a 2 − b2 I
JJ
=
ts
1
i a 2 − b2
H b K
ra
∴ By Cauchy’s Residue theorem,

F I
pi
I = 2πi (R) = 2πi GG 1

JJ = 2π
Hi 2
a −b 2
K a − b2
2
as
∴ z0
2π dθ
a + b sin θ
=
2π
a − b2
2
...(1)
g
If we replace a by 1 + a2 and b by – 2a, then

.c
z 2π dθ
=
2π
=
2π
=
2π
.
w
0 (1 + a 2 ) − 2a sin θ (1 + a ) − 4 a2 2 2 4
1 + a − 2a 2 1 − a2
Example 3. Use contour integration method to evaluate the following integral:

w
z π a dθ
, (a > 0).
w
0 a + sin 2 θ
2
Sol. I=
z
0
π
a2 +
(1
a dθ
− cos 2θ)
2
= 2a
z π
0 2
dθ
(2 a + 1) − cos 2θ
Put 2θ = φ, dθ =
dφ
2
=a z 2π
0 2
dφ
(2 a + 1) − cos φ
I = 2a z 2π
0
dφ
(4 a + 2) − ( e iφ + e − iφ )
2
...(1)
www.cgaspirants.com
www.cgaspirants.com
dz
But z = eiφ so that dφ = then (1) reduces to
iz
I = 2a
zC
(4 a 2
F
1
.
1 I iz
+ 2) – G z + J
H zK
dz
=
2a
i zC
dz
4a2 z + 2 z − z2 − 1
= 2ai zC 2
z − 2 z(1 + 2 a ) + 1
dz
2
= 2ai z C
dz
( z − α)( z − β)
where, α = (1 + 2a2) + 2a 1 + a 2 and β = (1 + 2a2) – 2a 1 + a 2

Clearly, |α|>1
m
∵ | αβ | = 1 ∴ |β|<1
∴ Only β lies inside C.
o
2ai 2ai 2 ai −i
.c
Residue (at z = β) is = Lt ( z − β) . = = =
z→β ( z − α)( z − β) β−α − 4a 1 + a 2
2 1 + a2
By Cauchy Residue theorem,
n
F −i I
ts
I = 2πi GG JJ = π
.
ra
H2 1+ a 2
K 1 + a2
Example 4. Apply Calculus of residues to prove that

pi
z 2π dφ
= 2
2 πa
where a > 0, b > 0, a > b.
as
2
0 (a + b cos φ ) (a − b 2 ) 3/2
z 2π dφ
z 2π dφ
g
Sol. Let, I= = ...(1)

0 (a + b cos φ) 2 0 RSa + b (e U
)V
2
.c
iφ – iφ
+e
T 2 W
w
dz
Put eiφ = z so that dφ = then,
iz
w
From (1), I= z RS 1
IJ UV
dz
= z − 4izdz
w
C b FG1
2
iz C (bz + 2az + b) 2
2
T a+
2
z+
Hz KW
=–
4i
b2 z FGH
C
z2 +
z dz
2az
+1
IJ
K
2
b
Poles are given by,
FG z 2 2az IJ 2
=0 ⇒ (z – α)2(z – β)2 = 0 where, α + β = –
2a
and αβ = 1.
H +
b
+1
K b
www.cgaspirants.com
www.cgaspirants.com
2a 4a2
− + −4
b b2 − a + a 2 − b2
Also, α= =
2 b
4a2 2a
− −4 –
b2 b − a − a 2 − b2
β= =
2 b
There are two poles, at z = α and at z = β each of order 2.
Since, | αβ | = 1
or |α||β|=1
But |β|>1 ∴ |α|<1 |∵ a>b
m
∴ Only z = α lies inside the unit circle | z | = 1.
Residue of f(z) at the double pole z = α is
o
LM R UVOP
.c
1 d − 4 iz
MN ST
= ( z − α) 2 . 2
(2 − 1) ! dz b ( z − α) 2 ( z − β) 2 WPQ z=α
=
LM d R − 4iz UOP
S
MN dz T b ( z − β) WPQ
n
2 V =–
b N
2 .M
z=α
ts
4i L ( z − β) . 1 − z . 2( z − β) O
( z − β) 2 PQ
2
4
z=α
ra
4i L (− β − z) O
=– 2 M
b N ( z − β) Q
P = – b4i ((−αα− −β)β) = b4i ((αα−+ββ))
3 2 3 2 3
z=α
pi
FG − 2a IJ
4i H b K = – ia
as
= .
b F2
GH b a − b IJK
2 3 2 2 3/2
2 (a − b )
2
g
I = 2πi M
L − ia OP = 2πa .
.c
∴
N (a − b ) Q (a − b )
2 2 3/ 2 2 2 3/ 2
w
z π cos 2θ dθ πp 2
w
= (0 < p < 1).

0 1 − 2p cos θ + p 2 1 − p2
z z
w
π cos 2θ dθ 1 2π cos 2θ dθ
Sol. I= 2 =
0 1 − 2 p cos θ + p 2 0 1 − p( e iθ + e − iθ ) + p2
=
1
2
real part of z 2π
0
e 2 iθ
(1 − pe iθ )(1 − pe − iθ )
dθ
=
1
2
real part of z C
(1 − pz) 1 −
z2
FG
H
p
z
IJ
K
dz
iz
writing e iθ = z, dθ =
dz
iz
=
1
2
real part of z C
− iz 2
(1 − pz)( z − p)
dz
www.cgaspirants.com
www.cgaspirants.com
=
1
2
real part of zC
f ( z) dz where, f ( z) =
− iz 2
(1 − pz)( z − p)
Poles of f(z) are given by (1 – pz)(z – p) = 0.
1
Thus z = and z = p are the simple poles. Only z = p lies within the unit circle C as
p
p < 1.
The residue of f(z) at z = p is
− iz 2 ip2
= lim (z – p) f(z) = lim (z – p) =−
z→ p z→ p (1 − pz)( z − p) 1 − p2
m
Hence by Cauchy’s residue theorem, we have
z f ( z) dz = 2πi × [Sum of residues within the contour]
o
C
F I = 2πp
.c
ip2 2
= 2πi −GH 1− p 2 JK 1 − p 2
which is purely real.
Hence, I=
1
2
real part of
n
zC
f ( z) dz =
πp2
ts
1 − p2
.
Example 6. Use Complex integration method to prove that

ra
z 2π sin 2 θ 2π
dθ = 2 (a − a 2 − b 2 ) , where 0 < b < a.
pi
0 a + b cos θ b
Sol. Let I=
z 2π sin 2 θ
z 2π 1 − cos 2θ
as
dθ = dθ
0 a + b cos θ 0 2(a + b cos θ)
z 2π 1 − e 2iθ
g
= Real part of dθ
0 2a + 2b cos θ
.c
1 FG
1 dz IJ
Put z = eiθ so that cos θ = z+
Hand dθ =
K
w
2 z iz
z 2π 1 − e 2iθ
z 1 − z2FG dz IJ =
z 1 − z2
w
Then
0 2a + 2b cos θ
dz =
C F 1 I H iz K
2a + b G z + J
C i(bz 2 + 2az + b)
dz
H zK
w
where C is the circle | z | = 1.

The poles of the integrand are the roots of bz2 + 2az + b = 0, viz.
− 2 a ± 4 a 2 − 4b2 − a ± a 2 − b2
z= =
2b b
− a + a 2 − b2 − a − a 2 − b2
Let α= and β =
b b
Clearly, | β | > 1 so that z = α is the only simple pole inside C.
www.cgaspirants.com
www.cgaspirants.com
Also, bz2 + 2az + b = b(z – α)(z – β)

Residue at z = α is
1 − z2 1 − z2 1 − α2
Lt ( z − α) . = Lt =
z→a ib( z − α)( z − β) z → α ib( z − β) ib(α − β)
FG 1 − αIJ
=
α
H α K = α(β − α) |∵ αβ = 1
ib(α − β) ib(α − β)
2 2
α a− a −b
=– =
ib ib2
m
∴ By Residue theorem,
z 1 − z2 a − a 2 − b 2 2π FH IK
o
dz = 2 πi . = 2 a − a 2 − b2
C i(bz 2 + 2az + b) ib2 b
.c
Hence I = Real part of
z 1 − z2 2π
dz = 2 a − a 2 − b2 . FH IK
ts
2
C i(bz + 2az + b) b
z 2π cos 2θ
n
Example 7. Using complex integration method, evaluate dθ .
0 5 + 4 cos θ
ra
(M.T.U. 2012, G.B.T.U. 2010)
z 2π e 2iθ
pi
Sol. Let I = Real part of dθ

0 5 + 2 ( e iθ + e − iθ )
as
= Real part of z
C 1
z2
5 + 2Gz + J
H zK
FG dz IJ
F I H iz K
writing e iθ = z
∴ dθ =
dz
g
iz
z
.c
1 z2
= Real part of 2
dz
i C 2z + 5z + 2
w
Singularities are given by

w
1
2z2 + 5z + 2 = 0 ⇒ z=– ,–2
2
w
1
z= − is the only pole which lies inside the unit circle C ≡ |z| = 1.
2
FG 1 IJ
H
Residue of f (z) at z = −
2
is
K
R=
FG z + 1 IJ . z 2
z2 1 1 FG IJ FG 2 IJ = 1
z→−
Lt
1
2
H 2 K i (2z + 1) (z + 2) = Lt
z→−
1
2
2i ( z + 2)
=
2i 4 H K H 3 K 12 i
Hence by Cauchy’s Residue theorem,
I= z
C
f ( z) dz = 2πi
FG 1 IJ = π .
H 12 i K 6
www.cgaspirants.com
www.cgaspirants.com
Example 8. Evaluate: z
0
π 1 + 2 cos θ
5 + 4 cos θ
dθ
Sol. Let I= z 0
2π 1 + 2 cos θ
5 + 4 cos θ
dθ = Real part of
z
0
2π 1 + 2e iθ
5 + 4 cos θ
dθ
= Real part of z
C
1 + 2z
5 + 2G z + J
H zK
FG dz IJ
F 1 I H iz K
Putting e iθ = z
∴ dθ =
dz
iz
= Real part of
1
i z
C 2
1 + 2z
2z + 5z + 2
dz
m
Poles are given by
1
(2z + 1) (z + 2) = 0 ⇒ z=– , – 2 (simple poles)
o
2
.c
1
z=– lies inside unit circle C ≡ | z | = 1
2
FG IJ FG IJ FG 1 IJ
Residue at z = −
H
1
2 z→−
K
= Lt z +
1
Hence by Cauchy’s Residue theorem,

1 1
.
1 + 2z
H
2 i (2 z + 1) ( z + 2)
n
2
=
K ts H 2i K z→ −
Lt
1
2
1 + 2z
z+2
=0
ra
I = 2πi (0) = 0
z2π 1 + 2 cos θ
dθ = 0
pi
0 5 + 4 cos θ
z π 1 + 2 cos θ
as
⇒ dθ = 0 | Using property of definite integrals

0 5 + 4 cos θ
z 2π
g
Example 9. Evaluate by Contour integration: e cos θ cos (sin θ – nθ) dθ.

0
.c
Sol. Let I= z 2π
e cos θ [cos (sin θ – nθ) + i sin (sin θ – nθ)] dθ
w
= z 2π
e cos θ . ei(sinθ–nθ) dθ =
z 2π iθ
e e . e − in θ dθ ...(1)
w
0 0
dz
w
Put eiθ = z so that dθ = then,

iz
Poles are given by

I= z C
ez .
1
zn
.
dz
iz
=–i zC z n+1
ez
dz
z=0 [of order (n + 1)]

It lies inside the unit circle.
R=
1 dn LM R|Sz n +1 − ie z U|VOP =
LM
− i dn
(e z )
OP =
−i
.
(n + 1 − 1) ! dz n MN T| z n +1 W|PQ z= 0
N
n ! dz n Q z=0
n!
www.cgaspirants.com
www.cgaspirants.com
∴ By Cauchy’s Residue theorem,
I = 2πi
FG − i IJ = 2π
H n !K n !
Comparing real parts, we have
z0
2π
e cos θ cos (sin θ – nθ) dθ =
2π
n!
.
Example 10. Evaluate the integral: z 0

π cos 2 3 θ
5 – 4 cos 2 θ
dθ (U.P.T.U. 2015)
1
z 2π cos 2 3θ 1
z 2π 1 + cos 6θ
m
Sol. Let, I= dθ = dθ ...(1)
2 0 5 – 4 cos 2θ 4 0 5 – 4 cos 2θ
Consider the integration round a unit circle c ≡ |z| = 1 so that z = eiθ
o
dz
.c
∴ dz = ieiθ dθ = iz dθ ⇒ dθ =
iz
FG IJ
Also, cos 2θ =
1 6 1
1 2iθ
2
n (e + e – 2iθ) =
FG
1 2
2 z
1
z + 2
IJ
ts
H K
and cos 6θ = z + 6
H K
ra
2 z
Then the given integral (1) reduces to
pi
F1 + z + 1 I
12
GH 2z JK dz 1
1
z 6
z z12 + 2 z6 + 1
as
I= ⋅ =– dz
4 c F
5 – 2G
z + 1 I iz
4 16i c FG 5
z5 z4 – z2 + 1
IJ
H z K J 2 H 2 K
g
1
.c
Singularities are, z = 0 (order 5), z = ± 2, ± (order 1)

2
w
1
Clearly, z = 0 and z = ± lie inside C.
2
w
1
Now we will find residues at z = 0 and z = ± .
w
z
12 6
+ 2z + 1
12
z + 2z + 1
6
LM1 – FG 5 z 2
–z
4 IJ OP –1
Let f(z) =
5 FG 4 5 2 IJ =
z5 N H2 KQ
z
H z – z +1
2 K
=
( z 6 + 1) 2 LM1 + 5 z 2
– z4 +
25 4
z + z 8 – 5 z 6 + ...
OP
z
5
N 2 4 Q
1
Residue of f(z) at z = 0 is the coefficient of in this laurent series expansion. Hence
z
25 21
R1 = Residue of f(z) at z = 0 = – 1 + =
4 4
www.cgaspirants.com
www.cgaspirants.com
1
R2 = Residue of f(z) at z =
2
= Lt.
FG z – 1 IJ . ( z 6 + 1) 2
z→
1 H 2K z 5
( z 2 – 2) z −
FG 1 IJ FG z + 1 IJ
2 H 2 K H 2K
( z6 + 1) 2 27
= Lt. =–
1 F
– 2) G z +
1 I 8
z5 ( z2
H 2 JK
z→
2
1
R3 = Residue of f(z) at z = –
m
2
= Lt
FG z + 1 IJ ⋅ ( z 6 + 1) 2
H 2K z FG IJ FG z + 1 IJ
o
z→ –
1 5 1
( z 2 – 2) z –
H K H 2K
.c
2
2
( z 6 + 1) 2 27
Now, by Cauchy-Residue theorem,

= Lt
z→ –
n 1
2 z5 ( z 2 – 2) z –
FG
H ts 1
2
IJ
K
=–
8
ra
1 π 21 27 3π FG IJ
I= −
16 i
[2π i (R1 + R 2 + R 3 )] = −
8 4
−
4
=
16
.
H K
pi
ASSIGNMENT
as
Evaluate the following integrals by using contour integration:
z z
g
2π dθ 2π dθ
1. (i) (U.K.T.U. 2011) (ii) (U.P.T.U. 2015)
.c
0 5 − 3 cos θ 0 2 + cos θ
zπ dθ
z π dθ
w
2. (i) (ii)
0 5 + 4 cos θ 0 17 − 8 cos θ
z
w
π a dθ
(iii) [G.B.T.U. (C.O.) 2011]
0 1 + 2a2 − cos 2θ
w
3. (i)
z
0
2π dθ
5 + 4 sin θ
(G.B.T.U. 2011) (ii)
z
0
π dθ
3 + sin 2 θ
(G.B.T.U. 2013)
4. (i)
z
0
2π dθ
(5 − 3 cos θ) 2
(ii)
z
0
π dθ
(a + cos θ) 2
5. (i)
z
0
π
2
a dφ
a + cos φ 2
(a > 0) (ii)
z
0
2π cos 2θ
1 − 2 p cos θ + p2
dθ , 0 < p < 1
6. (i)
z
0
2π cos 3θ
5 − 4 cos θ
dθ (U.P.T.U. 2009) (ii)
z
0
π cos 3θ
5 − 4 cos θ
dθ
www.cgaspirants.com
www.cgaspirants.com
(iii)
z 0
2π cos 3θ
5 + 4 cos θ
dθ (U.P.T.U. 2007) (iv)
z0
π cos 2θ
5 + 4 cos θ
dθ (U.P.T.U. 2014)
7.
z π a cos θ
− π a + cos θ
dθ ; a > 1 8.
z0
2π sin 2 θ
5 − 4 cos θ
dθ
9.
z0
2π
e − cos θ cos (nθ + sin θ) dθ ; n ∈ I
10. (i)
z0
2π
cos 2n θ dθ ; n ∈ I (ii)
z0
2π dθ
3 − 2 cos θ + sin θ
[G.B.T.U. 2013; U.P.T.U. 2014]
m
Answers
π 2π
1. (i) (ii)
o
2 3
.c
π π π
2. (i) (ii) (iii)
3 15 2 1 + a2
3. (i)
2π
3
5π
(ii)
2 3
π
n
πa
ts π 2πp2
4. (i) (ii) 5. (i) (ii)
ra
2 3/2
32 (a − 1) 1 + a2 1 − p2
π π π π
6. (i) (ii) (iii) − (iv)
pi
12 24 12 12
F I 2π
7. 2πa 1 − GG a
J 8.
π
9. (− 1) n
− 1 JK
as
n!
H a2 4
π (2n) !
10. (i) (ii) π.
g
2n − 1
(2) (n !) 2
.c
1.42.2. Integrals of the Type z ∞ f(x)

dx , where f(x) and F(x) are Polynomials in x
w
−∞ F(x)
x f(x)
such that → 0 as x → ∞ and F(x) has no Zeros on the Real axis.
w
F(x)
Consider the integral

f ( z)
z
dz over the closed contour C consisting of the real axis from – R
w
C F( z)
to R and the semi-circle CR of radius R in the upper half plane.
CR
–R O R X
www.cgaspirants.com
www.cgaspirants.com
f ( z)
We take R large enough so that all the poles of in the upper half plane lie within C.
F( z)
By residue theorem, we have
zC
f ( z)
F( z)
LM
dz = 2πi sum of the residues of
N
f ( z)
F( z)
in the upper half plane
OP
Q
zCR
f ( z)
F( z)
dz + z
f ( x)
F( x)
R
dx = 2πi sum of the residues of
–R
f ( z)
F( z)
LM
N
...(1) (∵ on the real axis, z = x)
OP
Q
If we put z = Reiθ in the first integral on the left side, then R is constant on CR and as z
moves along C1, θ varies from 0 to π.
z z
m
f ( z) π f (Re iθ )
∴ dz = Re iθ i dθ
CR F( z) 0 F(Re iθ )
o
z π f (Re iθ )
Re iθ idθ → 0
Rf (R)
.c
For large R, is of the order
0 F(Re iθ ) F(R)
∴
F(Re iθ )
Hence from (1), we have
z
0
π f (Re iθ ) n Reiθ idθ → 0 when R → ∞
ts
z
ra
∞ f ( x) LM
dx = 2πi sum of the residues of
f ( z)
OP
−∞ F( x) N F( z) Q
pi
EXAMPLES
as
Example 1. Using contour integration, prove that z

−∞
∞ dx
(1 + x 2 ) 2
=
π
2
.
g
z ∞ dx
.c
Hence or otherwise evaluate .

0 (1 + x 2 ) 2
z 1
w
Sol. Consider the integral f ( z) dz where f(z) =

taken round the closed contour
(1 + z 2 ) 2
C
w
C consisting of the semi-circle CR which is upper half of a large circle | z | = R and the part of
real axis from – R to R.
w
Y
For poles, (1 + z2)2 = 0
⇒ z2 = –1
⇒ z = ± i (Poles of order 2) CR
z = – i is outside C.
So z = i is the only pole inside C and is of order 2.
Residue of f(z) at z = i is
LM d R( z − i) 1 UVOP
MN dz ST
2
–R O R X
= .
( z − i) ( z + i) 2
2
WPQ z= i
L − 2 OP
=M =−
i
N (z + i) Q 3
z= i
4
www.cgaspirants.com
www.cgaspirants.com
By Cauchy’s residue theorem,

R
z dx
− R (1 + x )
Taking limit as R → ∞
2 2
+ zdz
2
C R (1 + z ) 2 = 2πi
−i
4
=
π
2
FG IJ
H K
∞
z dx
−∞ (1 + x 2 ) 2
+ Lim z dz
R → ∞ C R (1 + z 2 ) 2
=
π
2
...(1)
Now, z
CR
dz
(1 + z 2 ) 2
≤ z
C R |1 +
| dz|
z 2 |2
≤ z
CR
|dz|
{| z|2 − 1}2
m
∵ | z|= R on C R
= z π R dθ and| dz|= R dθ
o
0 (R 2 − 1) 2 also, 0 < θ < π
.c
πR
= → 0 as R → ∞
(R 2 − 1) 2
Hence,
∞
zdx
−∞ (1 + x 2 ) 2
=
π
2
n ts
∞
zdx
=
π ∞ dx π
= . z
ra
Now, 2 0 2 2 ⇒ 0 2 2
(1 + x ) 2 (1 + x ) 4
Example 2. Apply calculus of residues to prove that
z
pi
∞ dx π
2 2 2
= ; a > 0. (M.T.U. 2013)
0 (x + a ) 4a 3
z
as
Sol. Consider the integral f ( z) dz where Y

C
1
g
f(z) = taken round the closed contour C consist-

(a + z 2 ) 2
2
.c
CR
ing of the semi-circle cR which is upper half of a large circle
| z | = R and the part of real axis from – R to R.
w

w
(a2 + z2)2 = 0 –R O R X
i.e., 2 2
a +z =0
w
or z = ± ai each repeated twice.

The only pole within the contour is z = ai, and is of the order 2.
1
Here, f(z) = .
( z − ai) 2 ( z + ai) 2
1 d LM R 1 UVOP
MN ST
Residue (at z = ai) is = ( z − ai) 2 .
(2 − 1) ! dz ( z − ai) 2 ( z + ai) 2 WPQ z = ai
L d R 1 UOP
=M
L − 2 OP
MN dz ST ( z + ai) VWPQ 2
z = ai
=M
N (z + ai) Q
3
z = ai
−2 −1 1
= 3
= 3 3
= .
(2ai) 4a i 4 a3 i
www.cgaspirants.com
www.cgaspirants.com
Hence by Cauchy’s residue theorem, we have
z
C
f ( z) dz = 2πi (Sum of residues within C)
i.e., z
−R
R
f ( x) dx + z
CR
f ( z) dz = 2πi
FG 1 IJ
H 4a i K
3
or z R
−R
1
(a 2 + x 2 ) 2
dx + z CR
1
(a 2 + z 2 ) 2
dz =
π
2a3
...(1)
Now, zCR
1
(a + z 2 ) 2
2
dz ≤ z | dz|
C R | a + z 2 |2
2
z | dz|
m
≤
CR (| z|2 – a 2 ) 2
o
π R dθ
= , since z = Reiθ
0 (R 2 − a 2 ) 2
.c
πR
=
→ 0 as R → ∞
(R 2 − a 2 ) 2
Hence taking R → ∞, relation (1) becomes,
n ts
z ∞ 1
dx =
π
ra
2
2 2
−∞ ( a + x ) 2a3
z ∞ 1
dx =
π
pi
or 2 . 2 2
(a + x )
0 4 a3
as
z ∞
2 2
x2
2 2
−∞ (x + a )(x + b )
dx =
π
a+b
(a > 0, b > 0).
g
z z2
.c
Sol. Consider the integral f ( z) dz where f(z) =

taken round the
( z 2 + a 2 )( z 2 + b2 )
C
w
closed contour C consisting of the semi-circle CR which is upper half of a large circle | z | = R
and the part of real axis from – R to R.
w
z2
The poles of f(z) = are z = ± ia, ± ib. Of these, z = ia and z = ib lie in the
( z + a )( z 2 + b2 )2 2
w
upper half of the z-plane.

Residue of f(z) at z = ia is
z2
= lim ( z − ia)
z → ia ( z + a )( z 2 + b2 )
2 2
z2 a2 a
= lim 2 = – 2 = 2 .
z → ia z
( + ia)( z 2
+ b ) ia
2 (− a 2
b
+ ) 2i ( a – b2 )
Residue of f(z) at z = ib is
z2
= lim ( z − ib)
z → ib ( z + a )( z 2 + b2 )
2 2
www.cgaspirants.com
www.cgaspirants.com
z2 − b2 −b
= lim 2 2 = = .
z → ib ( z + a )( z + ib) (− b + a 2 )(2ib)
2
2i (a 2 − b2 )
By Cauchy’s residue theorem,
LM a − b OP
z−R
R
2
( x + a )( x + b ) 2
x2
2 2
dx + zCR 2 2
( z + a )( z + b )
z2
2 2
dz = 2πi
MN 2i (a − b ) 2i (a − b ) PQ 2 2 2 2
Taking limit as R → ∞
z ∞
2
x2
−∞ ( x + a )( x 2 + b 2 )
2
dx + Lt
R→∞ z CR
z
2
( z + a )( z + b ) 2
2
dz = π M
L a − b OP
Na − b Q
2 2 2 2
z ∞ x2
z z2 π
m
⇒ dx + Lt dz = ...(1)
−∞ ( x 2 + a 2 )( x 2 + b2 ) R→∞ CR 2 2 2 2
( z + a )( z + b ) a+b
z z2
z | z|2
o
Now, dz ≤ |dz|
CR ( z 2 + a 2 )( z 2 + b2 ) CR | z 2 + a 2 || z 2 + b 2 |
.c
≤ z | z|2
| dz|
=
n
R2
(R 2 − a 2 )(R 2 − b2 )
CR
z0
πts
(| z|2 – a 2 )(| z|2 – b2 )
R dθ | ∵ | z | = R on CR
ra
πR 3
= → 0 as R → ∞
(R 2 − a 2 )(R 2 − b2 )
z
pi
x2 ∞ π
∴ From (1), dx = .
−∞ ( x 2 + a 2 )( x 2 + b2 ) a+b
as
Example 4. (i) Apply Calculus of residues to prove that
z∞ dx π 2
g
44
= , (a > 0).
0 x +a 4a 3
.c
(ii) Using contour integration, evaluate z ∞ dx

1 + x4
. (U.P.T.U. 2007)
w
Sol. (i) Consider the integral z f ( z) dz where f(z) =

1
.
w
C z + a4
4
The poles of f(z) are given by

w
z4 + a4 = 0 ⇒ z4 = – a4 = a4eπi = a4e2nπi+πi
or z = ae(2n+1)πi/4 ; n = 0, 1, 2, 3.
Since there is no pole on the real axis, therefore, we may take the closed contour C
consisting of the upper half CR of a large circle | z | = R and the part of real axis from – R to R.
∴ By Cauchy’s residue theorem, we have
z R
−R
f ( x) dx + z CR
f ( z) dz = z C
f ( z) dz
or z R
−R
1
x4 + a4
dx + z
CR
1
z4 + a4
dz = 2πi ∑R +
...(1)
| where ∑R +
= sum of residues of f(z) at poles within C.
www.cgaspirants.com
www.cgaspirants.com
πi
The poles z = ae 4 and z = ae3πi/4 are the only two poles which lie within the contour C.
Let α denote any one of these poles, then
α 4 + a4 = 0 ⇒ α4 = – a 4 .
Residue of f(z) (at z = α) is = LM 1 OP =

1
=
α
MN dzd (z + a )PQ
3
4 4 4α − 4a4
z =α
1
∴ Residue at z = aeπi/4 is = − eπi/4
4 a3
1 e −πi/4
and residue at z = ae3πi/4 is = − e3πi/4 =
m
4a3 4 a3
1 LM e iπ/4
− e − iπ/4 1 OP i
π
∑R +
o
∴ Sum of residues = − =– i sin =− = .
2a 3
N 2 2a 3
Q 4 2 2 a3
.c
∴ From (1),
z −R
R
4
x +a
dx
4
+ zCR z +a4
dz
4
= 2 πi
n F –i I= π 2
GH 2 2 a JK 2a 3 3
ts ...(2)
Now,
z 1
dz ≤ z | dz|
≤ z | dz|
ra
4 4 CR | z 4 4 C R | z|4
CR z +a +a | −|a 4 |
= z π Rdθ
| ∵ |z | = R on CR
pi
0 R − a4
4
πR
= → 0 as R → ∞.
as
R − a4 4
Hence when R → ∞, relation (2) becomes

g
z ∞ dx π 2
z ∞ dx π 2
.c
4 4
= or = .
−∞ x +a 2a3 0 x +a44
4 a3
z
w
1
(ii) Consider the integral f ( z) dz where f(z) =
taken round a closed contour C,
1 + z4 C
w
consisting of the semi-circle CR which is upper half of a large circle | z | = R and the part of
w
1
The poles of f(z) = 4 are obtained by solving z4 + 1 = 0.
z +1
Now z4 + 1 = 0
⇒ z = (– 1)1/4 = (cos π + i sin π)1/4 = [cos (2nπ + π) + i sin (2nπ + π)]1/4
(2n + 1) π (2n + 1) π
= cos + i sin where n = 0, 1, 2, 3.
4 4
| By De Moivre’s theorem
π π 1 1
When n = 0, z = cos + i sin = +i
4 4 2 2
www.cgaspirants.com
www.cgaspirants.com
3π 3π 1 1
When n = 1, + i sin
z = cos =− +i
4 4 2 2
5π 5π 1 1
When n = 2, z = cos + i sin =− −i
4 4 2 2
7π 7π 1 1
When n = 3, z = cos + i sin = −i
4 4 2 2
π π
i 3i
Of these, only the poles corresponding to n = 0, 1, viz, z = e 4 and z = e 4 lie in the
upper half of z-plane.
π
π i
z−e 4 0
m
i
Residue of f(z) at z = e 4 is Lt 4
Form
i
π
z +1 0
z→ e 4
o
1
= Lt | By L’ Hospital’s rule
4 z3
.c
π
i
z→e 4
π
=
n
4e
1
3i
π
4
=
3i
1 −3i 4
4
π
e
−9 i
π
ts
ra
Similarly, residue of f(z) at z = e 4 is 41 e 4
1 –3iπ/4
Sum of residues = (e + e–9iπ/4)
4
pi
=
1
cos
3π
− i sin
3π LM
+ cos
9π
− i sin
9π OP
4 4 4 N 4 4 Q
as
1F
= G−
1
–
i
+
1
−
i IJ = − i 2
4H 2 2 2 2K 4
g
By Cauchy Residue theorem

.c
z R dx
+ z dz
= 2πi
F− i 2I = π 2
GH 4 JK 2
w
4 4
−R 1+ x CR 1+ z
Taking Limit R → ∞,
w
z ∞ dx
+ Lt z dz
=
π 2
...(1)
w
4 4
−∞ 1+ x R → ∞ CR 1+ z 2
Now, zCR
dz
1 + z4
≤ z
CR | z 4
| dz|
+ 1|
≤ zC R | z|4
| dz|
−1
=
R −1
πR
4
R
z 0
π
dθ |∵ | z | = R on CR
= → 0 as R → ∞
R4 − 1
∴ From (1), z−∞

∞
1+ x
dx
4
=
π 2
2
or z0
∞
Note. The above method can also be applied to some cases where f(x) contains trigonometric functions
1+ x
dx
4
=
π 2
4
.
also.
www.cgaspirants.com
www.cgaspirants.com
1.42.2. (a) Jordan’s Inequality

Consider the relation y = cos θ. As θ increases, cos θ decreases and therefore y decreases.
The mean ordinate between 0 and θ =
when θ = 0, ordinate is cos 0 i.e. 1

1
θ z0
θ
cos θ dθ =
sin θ
θ
Y
π sin π/2 2
when θ = , mean ordinate is i.e.
2 π/2 π P
Hence, when 0 < θ < π/2,
2 θ y
Mean ordinate lies between 1 and
m
π
2 sin θ O X
i.e., < <1
o
π θ
This is known as Jordan’s Inequality.
.c
1.42.2. (b) Jordan’s Lemma
If f(z) → 0 uniformly as | z | → ∞, then Lt

n R → ∞ CR z ts
e imz f(z) dz = 0, (m > 0)
where CR denotes the semi-circle | z | = R, I(z) > 0.

ra
Example 5. Apply calculus of residues to evaluate
z ∞ x sin x
dx , a > 0. (G.B.T.U. 2010)
pi
0 x2 + a2
Sol. Consider the integral z z e iz

f ( z) dz where f(z) =
taken round a closed contour
as
C z2 + a2
C, consisting of a semi-circle CR which is upper half of a large circle | z | = R and the part of
g
For poles, z2 + a 2 = 0
.c
⇒ z = ± ai
z = ai is the only pole which lie inside C.
w
z e iz ai e − a e − a
∴ Residue of f(z) at (z = ai) = lim ( z − ai) . = =
w
z → ai ( z − ai)( z + ai) 2ai 2

∴ By Cauchy Residue theorem,
w
z−R
R
2
x +a
x e ix
2
dx + z
CR
z e iz
2
z +a 2
dz = 2πi
F e I = πie
−a
GH 2 JK –a
Taking limit as R → ∞,
Lt
R → ∞ −Rz R
2
x +a
x e ix
2
dx + Lt z
R → ∞ CR
z e iz
z + a2
2
dz = πie–a
z
z
∞
−∞ 2
x e ix
x +a 2
dx + Lt
R → ∞ CRz z e iz
z + a2
2
dz = πie–a ...(1)
Since → 0 as | z | → ∞, therefore by Jordan’s Lemma,

z + a2
2
www.cgaspirants.com
www.cgaspirants.com
Lt
R → ∞ CR z ze iz
z + a2
2
dz = 0
∴ From (1),
x e ix
−∞ x 2 + a 2 z∞
dx = πi e–a
Comparing imaginary parts,
z∞
−∞
x sin x
x 2 + a2
dx = πe–a
or z
0
∞ x sin x
2
x +a 2
dx =
π −a
2
e .
m
Example 6. Evaluate by using Contour integration z ∞
0
cos ax
x2 + 1
dx ; a ≥ 0.
o
(U.P.T.U. 2006, G.B.T.U. 2011)
.c
Sol. Consider the integral z f ( z) dz where f(z) =
e iaz
taken round a closed contour C,
z2 + 1
C
ts
consisting of a semi-circle CR which is upper half of a large circle | z | = R and the part of real
axis from – R to R.
n
For poles, z2 + 1 = 0
ra
⇒ z=±i
z = i is the only pole which lies inside C.
pi
e iaz e−a
∴ Res. (z = i) = Lt ( z − i) . =
z→i ( z − i)( z + i) 2i
as
z
e iax R
dx + z e iax
dz = 2 πi
F I = πe
e−a
GH JK –a
g
−R x2 + 1 CR z2 + 1 2i
.c
Taking Limit as R → ∞,
z e iax
∞
dx + Lt
e iaz
dz = π e–a z ...(1)
w
2
−∞ x + 1 R → ∞ CR z 2 + 1
1
w
Since 2 → 0 as | z | → ∞, therefore by Jordan’s Lemma,

z +1
z
w
e iaz
Lt dz = 0 (a > 0)
R → ∞ CR z2 + 1
∴ From (1), z−∞

∞ e iax
x2 + 1
dx = π e–a
Equating real parts, we get
z ∞
−∞
cos ax
x2 + 1
dx = πe–a
or z 0
∞ cos ax
x2 + 1
dx =
π e− a
2
.
www.cgaspirants.com
www.cgaspirants.com
Example 7. Apply calculus of residues to prove that
z
∞ cosh ax
0 cosh πx
1 a
dx = sec , – π < a < π.
2 2
(M.T.U. 2014)
Sol. Consider
R + i and – R + i.
z c
e az
cosh πz
, c is the rectangle with vertices at – R, R,
Y
f(z) has simple poles given by
cosh πz = 0
or eπz + e–πz = 0
or eπz = – e–πz = e(2n+1)πi–πz – R+ i R+ i
(2n + 1)i i
m
whence, z = ; n = 0, ± 1, ± 2. Of these poles, only z =
2 2
lies inside c.
FG IJ LM e OP
o
i az
Residue at z =
H =
K M d (cosh πz)P
.c
2 –R O R
N dz Q
X
i
z=
ts
2
e ia / 2 e ia / 2 1 ia / 2
= = = e
n
iπ π πi
π sinh πi sin
2 2
ra
By residue theorem, we get
z z R
z 1
z −R
z 0
pi
f ( z) dz = f ( x) dx + f (R + iy) . idy + f ( x + i) dx + f (− R + iy) . idy

C −R 0 R 1
1 ia/2
as
= 2πi .
e = 2eia/2 ...(1)
πi
or I1 + I2 + I3 + I4 = 2eia/2.
z
g
1 e a( R + iy)
Now, | I2 | = idy
.c
0 cosh π (R + iy)
z
w
1 2e aR| e aiy ||i| dy

≤
0 | e π (R + iy) + e – π (R + iy) |
w
2 e aR dy
= = z2e aR
1
e πR − e − πR
→ 0 as R → ∞ | since – π < a < π.
w
0 e πR − e − πR
In the same way I4 → 0. Hence when R → ∞, we get from (1),
z−∞
∞ e ax
cosh πx
dx + z ∞
−∞ e a( x + i)
cosh π ( x + i)
dx = 2eia/2
or z
−∞
∞ e ax
cosh πx
dx − z ∞
−∞
e ax . e ai
− cosh πx
dx = 2eia/2 [∵ cosh π(x + i) = – cosh πx]
⇒ z −∞
∞(1 + e ia ) . e ax
cosh πx
dx = 2e ia / 2
⇒ z ∞
−∞
(e ia / 2 + e − ia / 2 ) e ax
cosh πx
dx = 2
www.cgaspirants.com
www.cgaspirants.com
or
e ax
−∞ cosh πx z
dx =
1
cos
a
2
∞
or
0 e ax
−∞ cosh πx z
dx +
∞ e ax
0 cosh πx
dx =z 1
cos
a
2
...(2)
Putting x = – t in the first integral, we get
∴
z
e ax
−∞ cosh πx
From (2),
0
dx = − z 0
∞
e − at
cosh πt
dt = z0
∞ e − ax
cosh πx
dx
m
∞ e ax + e − ax 1
dx =
0 cosh πx a
cos
o
2
z ∞ cosh ax 1 1 a
.c
or dx = = sec .
0 cosh πx a 2 2
2 cos
n
Example 8. Using contour integration, prove that:
2
ts z
0
∞ log (1 + x 2 )
1 + x2
dx = π loge 2.
ra
Sol. Consider the integral

C z log ( z + i)
z2 + 1
taken round a closed con-
pi
tour C which consists of semi-circle CR, the upper half of a large circle | z | = R and the part
of real axis from –R to R.
as
For poles, z2 + 1 = 0 ⇒ z = ± i
Only the pole z = i lies inside C.
g
π
log ( z + i) log (2i) log 2 + i
.c
2
Res. (z = i) = Lt ( z − i) . = =
z→i ( z − i)( z + i) 2i 2i
w
Lt z R log ( x + i)
dx + Lt z log ( z + i)
dz =
2 πi FG
H
log 2 + i
π IJ
πi
K FG IJ
w
= π log 2 + ...(1)
R→∞ −R 1 + x2 R → ∞ CR 1 + z2 2i 2 2 H K
LM OP
w
z log ( z + i) z log ( z + i)
Now, Lt = Lt .
z→∞ 2
z +1 z → ∞ z−i Nz+i Q
z log ( z + i)
= Lt Lt = 1.0 = 0
z→∞ z − i z→∞ z+i
Hence, Lt z
z → ∞ CR
z log ( z + i)
1 + z2
dz = 0
⇒ Lt z
z → ∞ CR
log ( z + i)
1 + z2
dz = 0
From (1), z ∞
−∞
log ( x + i)
1 + x2
FG
H
dx = π log 2 +
iπ
2
IJ
K
www.cgaspirants.com
www.cgaspirants.com
Equating real parts, we get
z 1 log ( x 2 + 1)
∞
−∞ 2 1 + x2
dx = π log 2
⇒ z
0
∞ log (1 + x 2 )
1 + x2
dx = π log 2.
ASSIGNMENT
Evaluate the following integrals using Contour integration:
z ∞ dx
z
∞ dx
m
1. (i) 2
(ii)
0 1+ x −∞ ( x + 1)3
2
z ∞ x2
z
∞ x2
o
(iii) 2 2
dx (G.B.T.U. 2012) (iv) dx ; a > 0
−∞ ( x + 1) −∞ ( x 2 + a2 ) 3
.c
2.
z ∞
2
dx
2 2 2
;a>b>0 3.
z
∞ x2
( x + 1)( x2 + 4)
2
dx (U.P.T.U. 2008)
ts
−∞ ( x + a )( x + b ) −∞
4.
z ∞ x2
dx 5.
z
∞ x2 − x + 2
dx
n
2 2 2
0 ( x + 9)( x + 4) −∞ x + 10 x2 + 9
4
z z x2
ra
∞ x ∞
6. (i) 2 2
dx (ii) dx
−∞ ( x + 1) ( x + 2 x + 2) −∞ ( x + 1) ( x 2 + 2 x + 2)
2 2
[U.P.T.U. (C.O.) 2009]

pi
7.
z ∞ x
dx 8.
z ∞ dx
as
2 2
−∞ ( x + 4 x + 13) −∞ x6 + 1
9. (i)
z ∞ cos mx
dx ; (m ≥ 0, a > 0) (ii)
z ∞ cos mx
dx ; m ≥ 0, a > 0
g
2 2
0 x +a 0 ( a2 + x 2 ) 2
.c
[U.P.T.U. (C.O.) 2008]
z ∞ x sin ax
z ∞ x sin πx
w
10. (i) dx ; a > 0 (ii) dx [U.P.T.U. (C.O.) 2009]

0 x +k2 2 −∞ x2 + 2 x + 5
z z
w
∞ cos x ∞ sin x
11. (i) 2 2 2 2
dx (a > b > 0) (ii) dx (a > b > 0)
0 (x + a ) (x + b ) 0 ( x2 + a2 ) ( x2 + b2 )
w
[U.P.T.U. (C.O.) 2008]
12.
14.
z−∞
∞ sin x
x2 + 4 x + 5
dx
If a > 0, prove that

13.
z
∞
0
3
x sin x
( x + a2 )( x 2 + b2 )
2
dx ; (a > 0, b > 0)
(i)
z ∞
−∞
a cos x + x sin x
x +a 2 2
dx = 2πe–a (ii)
z
∞
−∞
x cos x − a sin x
x2 + a2
dx = 0
eiz
[Hint: Consider f (z) = . At last multiply both Nr and Dr by x + ia and separate real and
z − ia
imaginary parts]
www.cgaspirants.com
www.cgaspirants.com
15. Prove that

z ∞
0
cos 2 x
(1 + x ) 2 2
dx =
π
2
FG
e
3
1+ 2
H
IJ LMHint: Take f ( z ) = 1 + e
K MN (1 + z
2iz
2 2
)
so that f ( x ) =
1 + cos 2x
2 2
(1 + x )
OP
PQ
2
16. By integrating e − z round the rectangle whose vertices are 0, R, R + ia, ia, show that
(i)
z 0
∞ 2
e − x cos 2ax dx =
e− a
2
2
π and (ii) z
0
∞ 2
e − x sin 2ax dx = e − a
2
z0
a 2
e y dy .
17. Apply calculus of residues to prove that

z 0
∞
sin x2 dx =
z ∞
0
cos x2 dx =
1
2
π
2
.
Answers
m
π π
1. (i) π/2 (ii) 3π/8 (iii) (iv)
2 8a3
o
π π 5π
2. 3. π/3 4. 5.
ab(a + b) 200 12
.c
π 7π π
6. (i) − (ii) 7. − 8. π/3
5 50 27
9. (i)
π −ma
2a
e (ii)
π e − am (am + 1)
n
4a 3 ts
10. (i)
π − ak
2
e (ii) – πe– 2π
Fe −b
e− a I
ra
π π
11. (i)
2(a 2 G
−b )H b 2
−
a JK (ii) 0 12. –
e
sin 2
π ( a 2e − a − b2e − b )
pi
13.
2 ( a 2 − b2 )
as
1.42.3. Integrals of the Type

f(x)
F(x) z ∞
dx , when F(x) has zeros on the real axis.
−∞
g
When the poles of f(z) lie on the real axis and also within the semi-circular region, then those
.c
which lie on the real axis can be avoided by drawing small semi-circles Cr , Cr′ etc. about those
poles as centres and small radii r and r′ in the upper half of the plane.
w
This method is said to be ‘indenting the semi-circular contour’.

When the semi-circle of radius R has been indented then f(z) is analytic along this
z
w
modified contour C and the integral f ( z) dz can be evaluated by Cauchy-Residue theorem.
z
C
w
sin mx ∞
Example. Evaluate dx , m > 0. [G.B.T.U. (C.O.) 2008, G.B.T.U. 2007]
x 0
Sol. Since sin mx is the imaginary part of eimx, we consider the function
e imz Y
. φ(z) =
z
This has a simple pole at z = 0, which lies on the C1
real axis. Enclose this singularity by a small semi-circle
C2 : | z | = r. Evaluate the function φ(z) over the contour
C2
C shown in the figure consisting of parts of the real axis
from –R to –r and r to R, the small semi-circle C2 and the
large semi-circle C1. Since the function has no singular- –R –r O r R X
ity within this contour, by Cauchy’s theorem, we have
www.cgaspirants.com
www.cgaspirants.com
z C
φ( z) dz = 0
⇒ z −r
−R
φ( x) dx + zC2
φ( z) dz + z
r
R
φ( x) dx + z
C1
φ( z) dz = 0
⇒
−R x z
e imx −r
dx +
e imz
C2 z
dz +
r
R e imx
x zdx +
e imz
C1 z
dz = 0 z...(1)
Substituting –x for x in the first integral and combining it with the third integral, we
z
get
z
z z
r
R e imx − e − imx
x
dx +
C2
e imz
z
dz +
e imz
C1 z
dz = 0
z z z
m
R sin mx e imz e imz
or 2i dx + dz + dz = 0 ...(2)
r x C2 z C1 z
o
Now
z z z e imz
dz =
1
dz +
e imz − 1
dz ...(3)
.c
C2 z C2 z C2 z
On C2 , z = reiθ
∴
z z z C2
1
z
dz =
0
π
n re iθ i dθ
re iθ
=−
π
0
idθ = – iπ ts
z z
ra
e imz − 1 | dz|
Also, dz ≤ M = πM
C2 z C2 | z|
pi
where M is the maximum value on C2 of | eimz – 1 | = | eimr (cos θ +i sin θ) – 1 |

Clearly, M → 0 as r → 0
z
as
e imz
∴ From (3), dz = − iπ
C2 z
g
Putting z = Reiθ in the integral over C1, we get ...(4)
z z z
.c
imz imR (cos θ + i sin θ)

e π e π
dz = iθ
Re iθ i dθ = i e imR cos θ . e–mR sin θ dθ
C1 z Re
w
0 0
Since, | eimR cos θ| ≤1
z z z
w
imz
e π π/2
∴ dz ≤ e − mR sin θ dθ = 2 e − mR sin θ dθ
C1 z 0 0
w
sin θ 2 π
Also, continually decreases from 1 to as θ increases from 0 to .
θ π 2
π sin θ 2 2θ
∴ For 0 ≤ θ ≤ , ≥ or sin θ ≥
2 θ π π
π
z e imz
z
dz ≤ 2 z 0
π/2
e −2 mR θ / π dθ = −
LM
N
π −2mRθ / π
mR
e
OP
Q
π/2
0
=
π
mR
(1 – e–mR)
As R → ∞, (1 – e–mR) → 0
mR
∴
e imz
C1 z
dz = 0z
www.cgaspirants.com
www.cgaspirants.com
Hence from (2), on taking the limit as r → 0 and R → ∞, we get

2i
∞ sin mx
x
dx − iπ = 0 z
z
0
∞ sin mx π
or dx = .
0 x 2
ASSIGNMENT
Apply calculus of residues to prove that:
1. (i) z ∞ xp − 1
1+x
dx =
π
sin π p
,0<p<1 (M.T.U. 2013) (ii) z ∞ xa − 1
1− x
dx = π cot a π, 0 < a < 1
m
0 0
(iii)
z ∞ x a −1
dx =
π
cosec
πa
,0<a<2
FG IJ
H K
o
2 2 2
0 1+ x
z z
.c
∞ cos x ∞ cos x π
2. (i) dx = 0 (ii) dx = sin a, (a > 0)
0 x −∞ a2 − x 2 a
3. (i)
z ∞ sin mx
2
x( x + a ) 2
dx =
2a
π
2
(1 – e–ma); a > 0 (ii)
z ∞
ts
sin πx
x(1 − x2 )
dx = π
n
0 0
z ∞ log x π
z ∞ (log x) 2 π2
ra
4. (i) 2 2
dx = − (ii) 2
dx =
0 (1 + x ) 4 0 1+x 8
z∞ xa FG IJ π πa
pi
5. (i) dx = (1 − a) sec ;–1<a<3

0 (1 + x 2 ) 2 H K 4 2
z xa F 2aπ IJ cosec (aπ); – 1 < a < 1

as
∞ 2π
(ii) dx = sin G
2
0 x − x +1 3 H 3K
z
g
∞ cos 2ax − cos 2bx

6. dx = π(b – a) if a ≥ b ≥ 0.
x2
.c
0
w
TEST YOUR KNOWLEDGE

w
1. Define analytic function and state the necessary and sufficient condition for function to be analytic.
(M.T.U. 2012)
w
2. If f(z) = u + iv is analytic, then show that the family of curves u(x, y) = c1 and v(x, y) = c2 are
mutually orthogonal. (M.T.U. 2012)
3. Using the Cauchy-Riemann equations, show that f(z) = | z |2 is not analytic at any point.
(M.T.U. 2013)
4. Find the constants a, b and c such that the function f(z) = – x + xy + y + i (ax + bxy + y2) is
2 2 2
analytic. (M.T.U. 2013)
5. Evaluate
z 1+i
0
z 2 dz. (U.P.T.U. 2008)
6.
7.
Evaluate the integral
z3 z
dz where C : | z | = 1.
Define isolated and non-isolated singular points.

C
e iz
(M.T.U. 2013)
(M.T.U. 2012)
www.cgaspirants.com
www.cgaspirants.com
8. Define removable and essential singular points with example. (M.T.U. 2012)
9. Define singular point of an analytic function. Find nature and location of the singularity of
z − sin z
f(z) = (M.T.U. 2013)
z2
10. Find the values of a and b for which the function f(z) = cos x (cosh y + a sinh y) + i sin x
(cosh y + b sinh y) is analytic.
11. If f(z) = u + iv is an analytic function and u = x2 – y2 – y then find its conjugate harmonic function
v(x, y).
12. If f(z) = u + iv is an analytic function and v = y2 – x2 then find its conjugate harmonic function u(x, y).
x 2 − y2
13. If u = is the real part of analytic function f(z) = u + iv then find f(z) in terms of z.
( x 2 + y2 ) 2
m
14. Evaluate
z C
dz
z−2
around the circle | z – 2 | = 4.
o
15. Evaluate (5z 4 − z 3 + 2) dz around the unit circle | z | = 1.
.c
C
16. If F(α) =
z C
5z 2 − 4 z + 3
z−α
ts
dz which C is the ellipse 16x2 + 9y2 = 144, then find F(2).
z
n
dz
17. Evaluate where C is | z – 3i | = 4.
C z2 + 9
ra
F z + 1I 3
18. Find residue of f(z) = GH z − 1 JK at z = 1.
pi
2z + 1
19. Find residue of f(z) = at the pole z = –1. (M.T.U. 2014)
z2 − z − 2
as
z2
20. (i) Find residue of f(z) = at the pole z = –1.
( z + 3 z + 2) 2
2
g
.c
z2
(ii) Find residue of f(z) = at the pole –1. (U.P.T.U. 2014)
z2 + 3 z + 2
w
z 4 − 3z
w
21. Evaluate dz where C is any simple closed path such that 1 ∈ C, 0 ∉ C.

C z2 − z
z − sin z
w
22. Find the nature of singularity of f(z) = at z = 0.

z3
23.
24.
Evaluate
z z−3
z 2 + 2z + 5
C
dz when C ≡ | z | = 1.
Let u(x, y) = 2x(1 – y) for all real x and y. Find a function v(x, y) so that f(z) = u + iv is analytic.
25. Let I =
z f (z)
C ( z − 1) ( z − 2)
dz where f(z) = sin
anti-clockwise. Find the value of I.

πz
2
+ cos
πz
2
and C is the curve | z | = 3 oriented
∞
26. Let ∑b
n=−∞
n z n be the Laurent’s series expansion of the function
1
z sinh z
, 0 < | z | < π, then find
b–2, b0 and b2.
www.cgaspirants.com
www.cgaspirants.com
27.
28.
Let f(z) =
15
∑z
n=0
n
for z ∈ . If : | z – i | = 2, then evaluate
z
( z − i )15
C
dz.
f (z)
Let u(x, y) be the real part of an entire function f(z) = u(x, y) + i v(x, y) for z = x + iy ∈ C. If C is the
positively oriented boundary of a rectangular region R in R2 then evaluate

z FGH
C
∂u
∂y
dx −
∂u
∂x
IJ
dy .
K
eiz
29. Consider the function f(z) = . Find the residue of f at the isolated singular point in the
z( z2 + 1)
upper half plane {z = x + iy ∈ : y > 0}.
30. Let S be the positively oriented circle given by | z – 3i | = 2. Then evaluate

z dz
.
m
2
S z +4
31. Let f(z) be an analytic function. Then evaluate

z 2π
f ( eit ) cos (t) dt.
o
0
.c
1 1
32. Let f(z) = then find the coefficient of in the Laurent’s series expansion of f(z)
2
z − 3z + 2 z3
33.
34.
for | z | > 2.
conjugate harmonic function v(x, y).

Define Harmonic function.
n ts
If u(x, y) = x3y – xy3 is the real part of analytic function f(z) = u(x, y) + i v(x, y), then find its
(M.T.U. 2014)
ra
Answers
pi
1 1 2 2
4. a=− , b = − 2, c = 5. − + i 6. –πi
2 2 3 3
9. removable singularity at z = 0 10. a = –1, b = –1 11. 2xy + x + c
as
1
12. 2xy + c 13. +c 14. 2πi 15. 0
z2
g
16. 30πi 17. π/3 18. 6 19. 1/3

.c
20. (i) – 4 (ii) 1 21. 2πi 22. removable singularity

23. 0 24. x2 – (y – 1)2 25. – 4πi
w
26. b–2 = 1, b0 = –1/6, b2 = 7/360 27. 2πi (1 + 15i) 28. 0

1 π
w
29. − 30. 31. πf ′(0) 32. 3

2e 2
33. x4 + y4 – 6x2y2 + c.
w
www.cgaspirants.com
www.cgaspirants.com
UNIT 2
Integral Transforms
2.1 INTRODUCTION
m
The theory of integral transforms affords mathematical devices through which solutions of
o
numerous boundary value problems of engineering can be obtained e.g. conduction of heat,
transverse vibrations of a string, transverse oscillations of an elastic beam, transmission lines
.c
etc.
ts
The choice of a particular transform to be employed for the solution of an equation
depends on the boundary conditions of the problem and the ease with which the transform can
be inverted. An integral transform when applied to a partial differential equation reduces the
n
number of its independent variables by one.
ra
2.2 DEFINITION
pi
The integral transform f(p) of a function F(x) is defined as
z b
as
I{F(x)} = f(p) = F(x) K(p, x) dx,

a
where K(p, x) is a known function of p and x, called the kernel of the transform : p is called the
g
parameter of the transform and F(x) is called the inverse transform of f(p).
Some of the well-known transforms are given below:
.c
(i) Laplace Transform. K(p, x) = e–px
z
w
∞
L{F(x)} = f(p) = F(x) e–px dx
0
w
(ii) Complex Fourier Transform. K(p, x) = eipx
z ∞
w
F{F(x)} = f(p) = F(x) eipx dx

−∞
(iii) Hankel Transform. K(p, x) = x Jn (px)
Hn{F(x)} = f(p) = z ∞
0
F(x) x Jn (px) dx
where Jn (px) is the Bessel function of the first kind and of order n.
(iv) Mellin Transform. K(p, x) = xp–1
M{F(x)} = f(p) = z
0
∞
f(x) xp–1 dx.
Other special transforms arise when the kernel is a sine or a cosine function. These lead
to Fourier sine or cosine transforms respectively.
131
www.cgaspirants.com
www.cgaspirants.com
2.3 FOURIER INTEGRAL THEOREM
Statement. If
(i) F(x) satisfies Dirichlet’s conditions in every interval (– c, c), however large.
(ii) z
−∞
∞
|F(x)| dx converges;
then
1 ∞ ∞
π 0 −∞
F(x) = zz
F (t) cos λ (t − x) dt dλ
The integral on the right hand side is called Fourier integral of F(x).
Proof. We know that Fourier series of a function F(x) in (– c, c) is given by
∑ FGH IJ
∞
m
a0 n πx n πx
F(x) = +
2 n=1
an cos
c
+ bn sin
c K ...(1)
o
1 c
where, F(t) dta0 =
.c
c −c
an =
1 c
F(t) cos
n πt
dt z
and bn =
c −c
1 c
c −c
F(t) sin
c
n πt
c
dt z
n
Substituting the values of a0, an and bn in (1), we get
ts
ra
F(x) =
1
2c z c
−c
F(t) dt +
1
c n=1 z
∑ LMNcos c cos c + sin c sin c OPQ F(t) dt
∞
n πx
c
−c
nπ t n πx n πt
pi
=
1
z c
F(t) dt +
1
∑z
∞ c
cos
nπ(t − x)
F(t) dt ...(2)
as
2c −c c n=1 −c c
If we assume that
z ∞
|F( x)| dx converges i.e., F(x) is absolutely integrable on the x-axis,
g
−∞
the first term on right side of (2) approaches 0 as c → ∞ since
z z
.c
1 c 1 ∞
F(t) dt ≤ |F(t)| dt
2c −c 2c
w
−∞
∑z
∞
1 ∞ nπ(t − x)
From (2), F(x) = Lim F(t) cos dt
w
c→ ∞ c n=1 −∞ c
z
w
∞
1 π
∑
∞
= Lim (Δλ) F(t) cos { n(Δλ )(t − x)} dt ...(3) where = Δλ
Δλ → 0 π −∞ c
n=1
This resembles a Riemann sum of a definite integral and is of the form
z
∞
∑ f (n Δλ)
∞
Lim i.e., f (λ) dλ
Δλ → 0 0
n=1
Hence as c → ∞, (3) reduces to
F(x) =
1
π zz
0
∞ ∞
−∞
F(t) cos λ(t – x) dt dλ ...(4)
which is known as Fourier integral of F(x). Eqn. (4) is true at a point of continuity.
www.cgaspirants.com
www.cgaspirants.com
INTEGRAL TRANSFORMS 133
At a point of discontinuity, the value of the integral on the right is

1
[F(x + 0) + F(x – 0)].
2
2.4 FOURIER SINE AND COSINE INTEGRALS
We know that
cos λ(t – x) = cos λt cos λx + sin λt sin λx
∴ Fourier integral of F(x) can be written as
F(x) = zz
1 ∞ ∞
π 0 −∞
F(t) {cos λt cos λx + sin λt sin λx} dt dλ
1 ∞
z z ∞ 1 ∞ ∞
z z
m
= cos λx F(t) cos λt dt dλ + sin λx F(t) sin λt dt dλ
π 0 −∞ π 0 −∞
...(1)
o
Case I. When F(x) is an odd function
F(t) cos λt is odd while F(t) sin λt is even. Thus the first integral in (1) vanishes and we get
.c
F(x) =
2
π z0
∞
sin λx z0
∞
ts
F(t) sin λt dt dλ
This is called Fourier sine integral.

n
Case II. When F(x) is an even function
ra
F(t) cos λt is even while F(t) sin λt is odd. Thus, the second integral in (1) vanishes and
we get
z z
pi
2 ∞ ∞
F(x) = cos λx F(t) cos λt dt dλ
π 0 0
as
This is called Fourier cosine integral.
2.5 COMPLEX FORM OF FOURIER INTEGRAL

g
.c
Fourier integral of F(x) is

1 ∞ ∞
zz
w
F(x) = F(t) cos λ (t − x) dt dλ ...(1)

π 0 −∞
Since cos λ(t – x) is an even function of λ, we have from (1),
zz
w
1 ∞ ∞
F(x) = F(t) cos λ (t − x) dt dλ ...(2)
2π −∞ −∞
w
Also, since sin λ(t – x) is an odd function of λ, we have

0= zz
1 ∞ ∞
2π −∞ −∞
F(t) sin λ (t − x) dt dλ
Multiplying (3) by i and adding to (2), we get
...(3)
F(x) =
1 ∞ ∞
zz
2π −∞ −∞
F(t) [cos λ(t – x) + i sin λ(t – x)] dt dλ
=
2π zz
1 ∞ ∞
−∞ −∞
F(t) . e iλ (t − x) dt dλ
F(x) = z z
1
2π
∞
−∞
e − iλx
∞
−∞
F(t) . e iλt dt dλ
which is known as the complex form of Fourier integral.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Express the function F(x) =

1 , for|x|< 1 UV
as a Fourier integral. Hence
0 , for|x|> 1 W
evaluate
0 zsin λ cos λx
∞
λ
dλ .
Sol. Fourier integral for F(x) is
F(x) =
1
π zz
0
∞ ∞
−∞
F(t) cos λ (t − x) dt dλ
1
zz ∞ 1 RS1, UV
− 1< t < 1
m
= cos λ (t − x) dt dλ ∵ F(t) =
π 0 −1 T0, otherwise W
1
z LMN sin λ (t − x) OP 1
o
∞
= dλ
π 0 λ Q
.c
−1
= zL
1 ∞ [sin λ (1 − x) − sin λ (− 1 − x)]
dλ
=
π 0
π 0
n
λ
λ
z MN
1 ∞ sin λ (1 + x) + sin λ (1 − x)
dλ ts OP
Q
2 ∞ sin λ cos λx
z
ra
= dλ
π 0 λ
z ∞ sin λ cos λx π π/2 , for|x|< 1 UV

pi
∴ dλ = F( x) =
0 λ 2 0 , for|x|> 1 W
1 π FG π IJ
as
At | x | = 1 i.e., x = ± 1, F(x) is discontinuous and integral has the value

2 2 H
+0 = .
4 K
z ∞ sin λ
z ∞ sin x
g
π π
Note. Putting x = 0, we get dλ = or, dx = .
0 λ 2 0 x 2
.c
Example 2. Using Fourier sine integral, show that
z FGH IJ RS UV
w
1 − cos πλ ∞ π/2, when 0 < x < π

λ
sin (xλ ) dλ =
0 0,K when x > π
.
T W
UV
w
π/2 , 0 < x < π

Sol. Let, F(x) = 0, x>π W
w
Using Fourier sine integral, we have
F(x) =
2
π z
0
∞
sin λx zL0
∞
F(t) sin λt dt dλ
=
2
π z∞
0
sin λx MNz
0
π π
2
sin λt dt dλ
OP
Q
= z ∞
0
FG − cos λt IJ dλ = FG 1 − cos πλ IJ sin (xλ) dλ
H λ K
sin λx
H λ K
π
0
z0
π
∴ z π
0
FG 1 − cos πλ IJ sin (xλ) dλ = F(x) = RSπ/2, 0 < x < πUV .
H λ K T 0, x > πW
www.cgaspirants.com
www.cgaspirants.com
Example 3. Using Fourier integral representation, show that
(i)
0 zω sin xω
∞
1 + ω2
π
dω = e − x , x > 0
2
Sol. (i) Fourier sine integral is
(ii) z ∞
0
cos xω
1+ω 2
π
dω = e − x , x ≥ 0.
2
F(x) =
2
π z z
∞
0
sin λx
0
∞
F(t) sin λt dt dλ
e–x =
2
π z z
0
∞
sin λx
0
∞
e − t sin λt dt dλ | Let F(x) = e–x
z LMN OP ∞
2 ∞ e−t
= sin λx (− sin λt − λ cos λt) dλ
π 0 1 + λ2 Q
m
0
=
2
π z FGH
∞
sin λx
λ
1+ λ2
dλ =
2
π
I
JK z ∞ λ sin λx
1 + λ2
dλ
o
0 0
z ∞ λ sin λx π −x
.c
⇒ 2
dλ = e ,x>0
0 1+ λ 2
or, z
1+ ω0
∞ ω sin ωx
π −x
2
(ii) Fourier cosine integral is
e ,x>0 2
dω =
n ts
z z
ra
2 ∞ ∞
F(x) = cos λx F(t) cos λt dt dλ
π 0
z z
0
2 ∞ ∞
pi
e–x = cos λx e − t cos λt dt dλ | Let F(x) = e–x

π 0 0
z LM e (− cos λt + λ sin λt)OP

−t ∞
as
2 ∞
= cos λx dλ
π 0
N1 + λ 2
Q 0
z F 1 I dλ
g
2 ∞
cos λx . G
=
H 1 + λ JK
.c
2
π 0
z
π −x ∞ cos λx ∞ cos ωx π
dω = e − x , x ≥ 0. z
w
⇒ e , x ≥ 0 or, dλ = 2
0 1+ λ2
2 0 1+ ω 2
Example 4. Using Fourier integral representation, show that
w
R| U|
z ∞ 0 , if x < 0
cos xω + ω sin xω
S| V|
w
dω = π/2 , if x = 0 .
1 + ω2
0
πe − x , if x > 0 T W
Sol. From example 3, we have
z0
∞ cos xω
1+ ω 2
π
dω = e − x , x > 0
2
and
z0
∞ ω sin xω
1+ ω 2
dω =
π −x, x > 0
2
e
Adding, we get
z ∞
0
1+ ω 2
dω =
π −x π −x
2
e + e ,x>0
= πe–x,
2
x>0
www.cgaspirants.com
www.cgaspirants.com
when x = 0, z
0
∞ cos xω + ω sin xω
1+ ω 2
dω = z ∞
0 1+ ω
dω
2
F
H
I
= G tan ωJ
K
−1
∞
0
=
π
2
when x < 0,
Hence,
z
0
∞ cos xω + ω sin xω
1+ ω 2
dω = z ∞
0
cos xω − ω sin xω
1 + ω2
dω = π e − x − π e − x = 0
2 2
R| U|
0 z
∞
1 + ω2
dω = π
0,
/2 ,
if x < 0
if x=0 .
πe − x, if x > 0
S|
T
V|
W
Example 5. Find the complex form of the Fourier integral representation of
m
f(x) = S
Re ,
x > 0 and k > 0 − kx
T 0,
o
otherwise
Sol. We know that the complex form of Fourier integral representation of f(x) is given by
.c
f(x) =
1
z z
∞
e − iλx
∞
f (t) e iλt dt dλ ...(1)
Here, f(t) =
n 2π
RSe ,
− kt
T 0,
−∞ −∞
x > 0 and k > 0

otherwise
ts
z z LM e e OP
ra
1 ∞ ∞
∴ From (1), f(x) = e − iλx − kt iλt
dt dλ
2π −∞ N0 Q
z z
pi
=
1 ∞
e − iλx
LM e∞
−( k − iλ )t O
dtP dλ
2π −∞ N
0 Q
as
=
1
2π z ∞
e − iλx
LM e OP
− ( k − iλ ) t
N − (k − iλ) Q
∞
dλ =
1
2π z ∞ e − iλx
k − iλ
dλ .
g
−∞ −∞
0
.c
ASSIGNMENT
w
1. Using Fourier integral representation, show that
z
w
2a ∞ cos λx
e–ax = dλ ; a > 0, x ≥ 0.
π 0 λ2 + a2
w
2. Find Fourier sine integral for F(x) = e–αx.

3. Using Fourier integral representation, show that
2 z 2
∞
2
0 (λ + α )(λ + β )
2
π ( e − αx − e − β x )
λ sin λx
2 (β2 − α 2 )
Hence find the Fourier sine integral representation of e–x – e–2x.
.
dλ =
0, R|x<0 U|
4. S| V|
If F(x) = x, 0 ≤ x ≤ π , then show that
0, T
x>π W
F(x) =
1
πλ2 z ∞
0
[λπ sin λ(π – x) + cos λ(π – x) – cos λx] dλ
www.cgaspirants.com
www.cgaspirants.com
5. Using Fourier integral formula, prove that
(i)
z 0
∞ F λ + 2 I cos λx dλ = π e
2
GH λ + 4 JK
4 2
−x
cos x, if x > 0
(ii) z 0
∞ F λ I sin λx dλ = π e
4
3
GH λ + 4 JK 2
−x
cos x, if x > 0.
6. Using Fourier integral method, prove that
(i) z FGH
0
∞ sin πλ
1– λ K
I sin λx dλ = R|S π sin x,
J 2
|T 2 0,
if 0 ≤ x ≤ π
if x > π |W
U|
V
F πλ I
cos G J cos λx R| π cos x, πU
z H 2K if | x | < |
m
∞
π V.
(ii) dλ = S 2 2
|T 0, if | x | > |
2
0 1− λ
2W
o
7. Find Fourier sine integral representation of
.c
0, 0 < x < 1R| U|
S|
(i) F(x) = k, 1 < x < 2 , where k is a constant. V| (ii) F(x) = x2, 0 ≤ x ≤ 1.
8.
0, x>2 T
Find Fourier cosine integral representation of
RSsin x, 0≤x≤π
W
UV0≤ x≤2
n ts RSx, UV
ra
(i) F(x) = (ii) F(x) =
T 0, x>π W
otherwise T0, W
9. Find the complex form of the Fourier integral representation of
R|0,
pi
−∞<x<−1
F(x) = S x, − 1< x < 0
|T0, x>0
as
10. If Fourier cosine integral of a function F(x) can be represented as
z ∞ 2
z ∞
g
F(x) = A(λ) cos xλ dλ where A(λ) = F(t) cos λt dt

0 π 0
z
.c
1 ∞
A
FG λ IJ cos xλ dλ.
Prove that F(ax) =
a H aK (a > 0)
w
Answers
w
2.
2
z ∞ λ sin λx
dλ 3.
6
z ∞ λ sin λx
dλ
w
2 2
π 0 α +λ π 0 (λ2 + 1)(λ2 + 4)
7. (i) F(x) =
2k
π z FGHLF
0
∞ cos λ − cos 2λ
λ
IJ sin λx dλ
K
(ii) F(x) =
2
π z MNFGH
0
∞ −1 2
λ
+ 3 cos λ +
λ
IJ
2 sin λ
K λ2
2
− 3
λ
OP
Q sin λx dλ
8. (i) F(x) = –
2
π z GH ∞
0
1 + cos λπ
2
λ −1 K
I cos λx dλ
J
(ii) F(x) =
2
π z FGH
0
∞
2 sin 2λ cos 2λ − 1 I
λ
+
λ
JK cos λx dλ
2
9. F(x) =
1
2π z−∞
∞
e–iλx
RS 1 + FG 1 − 1 IJ e UV dλ.
T λ H iλ λ K W
2 2
− iλ
www.cgaspirants.com
www.cgaspirants.com
2.6 COMPLEX FOURIER TRANSFORM
If a function F(x) defined on the interval (– ∞, ∞) is piecewise continuous in each finite partial
interval and absolutely integrable in (– ∞, ∞), then
f(p) = F{F(x)} = z
−∞
∞
F(x) . eipx dx
is defined as the Fourier transform of F(x) and is denoted by f(p). The function F(x) is called
the inverse Fourier transform of f(p).
The inverse formula for Fourier transform is given by
F–1 {f(p)} = F(x) =

1
2π z ∞
−∞
f(p) . e–ipx dp.
m
Note. Some authors write the formulae as
f(p) =
1
z ∞
eipx . F(x) dx and F(x) =
1
z ∞
e–ipx . f(p) dp.
o
2π −∞ 2π −∞
.c
2.7 FOURIER SINE TRANSFORM
ts
The infinite Fourier sine transform of the function F(x), 0 < x < ∞ is denoted by Fs {F(x)} or fs(p)
and defined by
z
n
∞
fs(p) = F(x) . sin px dx
ra
0
and the function F(x) is called the inverse Fourier sine transform of Fs {F(x)}.
The inverse formula is given by
pi
F(x) = Fs–1 {fs (p)} =

2
π z ∞
0
fs (p) . sin px dp.
as
2.8 FOURIER COSINE TRANSFORM

g
The infinite Fourier cosine transform of the function F(x). 0 < x < ∞ is denoted by Fc{F(x)} or
.c
fc(p) and defined by
fc(p) = Fc{F(x)} = z ∞
F(x) . cos px dx
w
0
and the function F(x) is called inverse Fourier cosine transform of fc(p).
w
The inverse formula for infinite Fourier cosine transform is given by

2
z ∞
w
F(x) = Fc–1 [fc(p)] = fc (p) . cos px dp.

π 0
2.9 SOME IMPORTANT RESULTS
1. z0
∞ sin mx
x
π
dx = , (m > 0)
2
2. z
0
∞ 2
e − x dx =
2
π
3. z
e +e∞ ax
0 e πx + e − πx
− ax
1 a
dx = sec , (– π < a < π)
2 2
4. z
0
∞ e ax − e − ax
e πx − e − πx
1 a
dx = tan , (– π < a < π)
2 2
5.
∞
z sin rx
−∞ ( x − b) 2 + a 2
π
dx = e − ar sin br, (r > 0).
a
www.cgaspirants.com
www.cgaspirants.com
2.10 PROPERTIES OF FOURIER TRANSFORMS
2.10.1. Linear Property

If f1(p) and f2(p) are the Fourier transforms of F1(x) and F2(x) respectively then
F [c1F1(x) + c2F2(x)] = c1f1(p) + c2f2(p)
where c1 and c2 are constants.

Proof. We have,
F [c1F1(x) + c2F2(x)] = z ∞
−∞
[ c1F1 ( x) + c2 F2 ( x)] . e ipx dx
z z
m
∞ ∞
= c1 F1 ( x) . e ipx dx + c2 F2 ( x) eipx dx
−∞ −∞
o
= c1 F [F1(x)] + c2F[F2(x)] = c1 f1(p) + c2 f2(p).
.c
2.10.2. Change of Scale Property (Similarity Theorem)
If f(p) is the complex Fourier transform of F(x) then
F [F(ax)] =
1
a
f
n FG pIJ
H aK , a ≠ 0.ts
ra
Proof. F[F(ax)] = z ∞
−∞
e ipx . F(ax) dx
Put ax = t
⇒ dx =
dt
a
pi
z
F pI
= ez ∞
F(t)
dt = 1 ∞ e i GH a JK t F(t) dt = 1 f p .
ip
t
a
a a
FG IJ
H K
as
−∞ a a −∞
From this property, it is evident that if the width of a function is decreased while its
height is kept constant then its Fourier transform becomes wider and shorter. If its width is
g
increased, its transform becomes narrower and taller.

.c
Remark. If fs(p) and fc(p) are Fourier sine and cosine transforms of F(x) respectively then
1 FG p IJ 1
fc
FG p IJ .
H aK
w
Fs [F(ax)] = fs and Fc [F(ax)] =

a H aK a
2.10.3. Shifting Property

w
If f (p) is the complex Fourier transform of F(x) then

w
F [F(x – a)] = eiap f(p)
Proof. F [F(x – a)] = z−∞

∞
F( x − a) . e ipx dx
Put x − a = u
∴ dx = du
= z∞
−∞
F(u) e ip(u + a) du = eiap
z ∞
−∞
F(u) e ipu du = eiap f(p).
2.10.4. Modulation Theorem

If f(p) is the complex Fourier transform of F(x) then
1
F [F(x) cos ax] = [f(p + a) + f(p – a)]
2
www.cgaspirants.com
www.cgaspirants.com
Proof. F [F(x) cos ax] = z −∞

∞
e ipx . F( x) cos ax dx
= z ∞
−∞
e ipx . F( x) .
Fe
GH
iax
+ e − iax
2
dx
I
JK
1 ∞ i ( p + a) x
2 −∞
1
=e
LM
z
. F( x) dx +
N
∞
−∞
e i( p− a) x . F( x) dx z OP
Q
= [f(p + a) + f(p – a)].
2
This result has application in radio and television where the harmonic carrier wave is
modulated by an envelope.
m
Note. If fs(p) and fc(p) are Fourier sine and cosine transforms of F(x) respectively then
1
(i) Fs [F(x) cos ax] = [f (p + a) + fs(p – a)]
o
2 s
1
.c
(ii) Fs [F(x) sin ax] = [f (p – a) – fc(p + a)]
2 c
1
(iii) Fc [F(x) sin ax] =
2.10.5. Convolution Theorem

[f (p + a) – fs(p – a)].
2 s
n ts
[M.T.U. 2014, U.P.T.U. 2010]
The convolution of two functions F(x) and G(x) over the interval (– ∞, ∞) is defined as
ra
F∗G= z
−∞
∞
F(u) . G(x – u) du = H(x).
pi
Statement. The Fourier transform of the convolution of F(x) and G(x) is the product of their
Fourier transform i.e.,
as
F{F(x) ∗ G(x)} = F{F(x)} . F{G(x)}

g
Proof. We have,
RS
z UV
.c
∞
F{F(x) ∗ G(x)} = F F(u) . G ( x − u) du
T −∞ W
w
= z z ∞ RS
T
∞
F(u) . G ( x − u) du e ipx dx
UV
W
w
−∞ −∞
z RSTz
∞ ∞
G ( x − u) . e ipx dx du
UV on changing the order
w
= F(u)
−∞ −∞ W of integration
= z RSTz
∞
−∞
F(u) .
∞
−∞
UV
e ip( x − u) . G ( x − u) d( x − u) e ipu du
W
= z RSTz
∞
−∞
e ipu F(u)
∞
−∞
e ipt G (t) dt du
UV
W | where x – u = t (say)
= z ∞
−∞
e ipu F(u) du . F{G(t)}
= z ∞
−∞
e ipx . F( x) dx . F{G( x)} = F{F(x)} . F{G(x)}.
www.cgaspirants.com
www.cgaspirants.com
2.10.6. If Fs [F(x)] = fs(p) and Fc [F(x)] = fc(p), then

d d
(i) Fs [x F(x)] = – [f (p)] (ii) Fc [x F(x)] = [f (p)].
dp c dp s
Proof. (i) fc(p) = z 0

∞
F( x) cos px dx
d
{f (p)} = –
dp c z 0
∞
F( x) . x sin px dx = – Fs [x F(x)]
d
⇒ Fs [x F(x)] = – [f (p)]
dp c
m
∞
(ii) fs(p) = F( x) . sin px dx
0
d
z ∞
o
{f (p)} = F( x) . x cos px dx = Fc [x F(x)]
dp s 0
.c
d
⇒ Fc [x F(x)] = [f (p)]
dp s
n ts
EXAMPLES
ra
Example 1. Find the Fourier transform of following functions:

U|
pi
e iωx , for a < x < b

V| 1
, |x|≤ ε
U|
(i) F(x) =
0, otherwise W
(ii) F(x) =
2ε V|
W
as
0 , x>ε
(iii) F(x) = e–|x| (iv) F(x) = e–a|x|, a > 0
t, for|t|< a UV
g
(v) F(t) = . (U.P.T.U. 2008, 2009)

0, for|t|> a W
.c
Sol. (i) f(p) = z ∞

F( x) . e ipx dx = z b
e iωx . e ipx dx
w
−∞ a
L e OP
= M
i(ω + p) x b
e i(ω + p)b − e i(ω + p) a
w
= .
N i(ω + p) Q a
i(ω + p)
z
w
∞
(ii) f(p) = F( x) . e ipx dx
−∞
= z−ε
ε 1 ipx
2ε
e dx =
1
2∈ z −ε
ε
(cos px + i sin px) dx
=
1
2ε
.2 z
0
ε
cos px dx =
1 sin px
ε p
FG
H
IJ
K
ε
0
=
sin pε
pε
.
(iii) f(p) = z ∞
−∞
F( x) e ipx dx = z−∞
∞
e–|x| eipx dx
= z0
−∞
e x e ipx dx + z 0
∞
e–x eipx dx
www.cgaspirants.com
www.cgaspirants.com
L e OP
=M
(1+ ip) x 0 Le
+M
OP
− (1− ip) x ∞
=
1
+
1
=
2
.
N 1 + ip Q −∞ N − (1 − ip) Q 0
1 + ip 1 − ip 1 + p2
(iv) We have just proved that
2
F{e–|x|} =
1 + p2
using change of scale property, we get
F{e–a|x|} =
1 LM 2 OP = 2a
.
a MM 1 + FG pIJ 2
PP p2 + a 2
N H aK Q
z z
m
∞ a
(v) f(p) = F(t) . eipt dt = t eipt dt
−∞ −a
zL z
o
a a
= t (cos pt + i sin pt) dt = 2i t sin pt dt
.c
−a 0
R F cos pt IJ UV − 1 . FG − cos pt IJ dtOP

z
a
= 2i MSt . G −
a
MNT H p K W
L a
n H p K PQ
1 F sin pt I O
0
ts
0
a
= 2i M− cos ap + G JP
MN p pH p K P
ra
Q 0
L a
= 2i M− cos ap +
1 O
sin apP =
2i
pi
(sin ap – ap cos ap).

N p p Q p 2 2
Example 2. Find the Fourier transform of the function

as
sin ax
(i) F(x) = xe–a|x|, a > 0 (ii) F(x) = ,a>0
x
g
R|1 + x , for − a < x < 0

U|
.c
| a
(iii) F(x) = S1 − x ,
|V .
|| a for 0 < x < a
||
w
T 0, otherwise W
z
w
∞
Sol. (i) f(p) = x e–a| x | eipx dx
z z
−∞
z z
w
0 ∞ 0 ∞
= x eax eipx dx + x e–ax eipx dx = x e(a+ip)x dx + x e–(a–ip)x dx
−∞ 0 −∞ 0
R x e UV − 1 . e dx + RSx . e
z UV − 1 . e
z
( a + ip) x 0 ( a + ip) x − ( a − ip) x ∞ − ( a − ip) x
0 ∞
=S
|T a + ip |W a + ip
−∞
|T − (a − ip) |W
−∞ − (a − ip)
dx
0
0
=– M
Le OP + LM e
( a + ip) x 0 OP = − 1 + 1 = 4 iap − ( a − ip) x ∞
N (a + ip) Q N − (a − ip) Q (a + ip) (a − ip) (a + p )

2 2 2 2 2 2 2
z
−∞ 0
(ii) f(p) =
−∞
sin ax ipx
∞
x
e dx =
∞ sin ax
x
(cos px + i sin px) dx z
z RST
−∞
=2
0 z
∞ sin ax
x
cos px dx =
0
∞ sin ( a + p) x
x
+
sin (a – p) x
x
dx
UV
W ...(1)
www.cgaspirants.com
www.cgaspirants.com
Case I. If |p| < a then a + p and a – p both are positive and then (1) gives
π π
+ = π. f(p) =
2 2
Case II. If |p| > a then for positive values of p, (a + p) is positive whereas (a – p) is
negative and for negative values of p, (a + p) is negative while (a – p) is positive. Consequently,
we get
π π π π
f(p) = − or – + = 0
2 2 2 2
Hence, f(p) =
RSπ, | p|< a UV
T0, | p|> a W
z z z FGH
m
∞ 0 FG 1 + x IJ e a IJ
x ipx
(iii) f(p) =
−∞
F(x) . eipx dx =
−a H aK ipx dx +
0
1−
aKe dx
z z z
o
a 1 0 1 a
= eipx dx + x eipx dx – x eipx dx ...(1)
.c
−a a −a a 0
1 LMF e I OP LMF e I OP
F e I
z z
ipx a ipx 0 a
0 e ipx ipx
e ipx
ts dx – 1
a
=G
H ip JK a MNGH x . ip JK
+
−a −a
−
−a
1.
ip a PQ MNGH x . ip JK 0
−
0
1.
ip
dx
PQ
n
F I 1 LM e F I OP − 1 LMa e F I OP
ipa − ipa − ipa 0 a
e −e 1 e ipx ipa
1 e ipx
=G
H ip JK a MNa ip GH JK P a M ip GH JK P
ra
+ − − .
ip ip
Q N −a
ip ip
Q 0
1L 1 O
pi
1
= M − 1)P
− ipa ipa
(1 − e )− (e
a Np 2
p 2
Q
as
2 1 2
= 2
− 2 [e–ipa + eipa] = (1 – cos ap) ; p ≠ 0
ap ap ap2
g
when p = 0,
z z z
.c
a 1 0 1 a
f(p) = 1 . dx + x dx − x dx | From (1)
−a a −a a 0
w
1 x2 F I 0
1 x2 F I a
a a
= 2a + GH JK − GH JK = 2a – − = 2a – a = a
w
a 2 −a
a 2 0
2 2
R| 2 U|
w
(1 − cos ap) ; p≠0

Hence, f(p) = S ap
|Ta;
2
V| .
p = 0W
Example 3. Find the complex Fourier transform of dirac delta function δ(x – a).
Sol. F{δ(x – a)} = z−∞

∞
δ(x – a) . eipx dx
z F I a+ h
a+ h 1 ipx 1 e ipx
= Lim
h→0 a h
e dx = Lim
h→0 h ip GH JK a
F e − 1I = e
iph F e − 1I = 1
iph
= Lim e ipa
h→0
GH iph JK ipa ∵ Lt
iph → 0
GH iph JK
Remark. For the function δ(t), F[δ(t)] = 1.
www.cgaspirants.com
www.cgaspirants.com
2
Example 4. Find the Fourier transform of e − x . Hence find the Fourier transform of
2
− ax 2
(i) F(x) = e , (a > 0) (ii) F(x) = e − x /2
2 2
(iii) F(x) = e −4(x − 3) (iv) F(x) = e − x cos 2x.
Sol. f(p) = z−∞

∞ 2
e − x . e ipx dx =
R|FG x − ip IJ U|
z ∞
−∞
e−( x
2
− ipx)
dx
z
2
p2
∞ − S|H 2 K + V|
= e T 4 W dx
−∞
z
FG ip IJ 2
Put x −
ip
=z
=e
2
− ( p / 4)
∞
e
H
− x−
2 K dx 2
m
−∞ ⇒ dx = dz
= e −( p
2
/ 4)
z ∞
e − z dz
2
o
z
−∞
∞
− x2 − ( p2 / 4 ) 2 2
e − z dz = π e − ( p / 4)
.c
⇒ F( e ) = 2e ...(1)
0
(i) By change of scale property,
F(e − ax ) =
2 1
a
n
πe
−
1
4
FG p IJ
H aK
2
=
a
e ts
π − ( p 2 / 4 a)
1
ra
(ii) Comparing with the result of deduction (i), we get a =

2
2 2
Hence, F(e − x / 2 ) = 2π e − ( p / 2)
pi
(iii) We have, from (1),

as
2 2
F (e − x ) = π e − ( p / 4)
1 F pI 2
2 π − 4 GH 2 JK
2 π −( p2 / 16)
∴ F ( e −4 x ) = F { e –( 2 x ) } =
g
e = e | By change of scale property

2 2
.c
2 π 3ip –( p2 / 16) π [3ip− ( p2 / 16)]

Hence, F { e −4 ( x − 3) } = e e = e | By shifting property
2 2
w
(iv) We have, from (1),

2 2
F(e − x ) = π e − ( p / 4)
w
2 π 1
− ( p + 2) 2
1
− ( p − 2) 2 LM OP
∴ F( e − x cos 2x) =
e 4 +e 4 .
N | By Modulation theorem
Q
w
2
Example 5. Find the inverse Fourier transform of f(p) = e–|p|y.
Sol. F(x) =
1 ∞ −| p|y − ipx
2π −∞
e .e dp z
=
1
2π −∞
0
e py e − ipx dp +
LM
N
L
0
∞
z
e − py e − ipx dp
O
OP
Q z
1 MR e UV + RS e UV P
p( y − ix ) 0 − ( y + ix) p ∞
= S
2π M|T y − ix |W |T − ( y + ix) |W PQ
N −∞ 0
1 L 1 1 O y
= M +
2π N y − ix y + ix QP =
π( y + x )
. 2 2
www.cgaspirants.com
www.cgaspirants.com
Example 6. Find the Fourier transform of
F(x) =
1, |x|< a
.
UV (U.P.T.U. 2014, 2010, 2008)
0, |x|> a W
Hence evaluate
(i) z
−∞
∞ sin ap cos px
p
dp (ii) z0
∞ sin p
p
dp .
Sol. f(p) = z ∞
−∞
2 sin ap
F( x) . e ipx dx = z a
−a
1 . e ipx dx = 2 z0
a
cos px dx
= ,p≠0
p
m
For p = 0, we find f(p) = 2a
Taking Inverse Fourier transform of f(p), we get
z z
o
1 ∞ 1 ∞ 2 sin ap − ipx
F(x) = f ( p) . e − ipx dp = e dp
.c
2π −∞ 2π −∞ p
=
1
π z ∞
−∞
sin ap (cos px − i sin px)
p
dp
ts
RS1, UV z
n
| x| < a 1 ∞ sin ap cos px
⇒ = dp | Second integral vanishes
T0, | x| > a π W −∞ p
ra
⇒
−∞ z
sin ap cos px
p
∞
dp =
π, | x|< a
0, | x| > a
RS
T
UV
W ...(i)
pi
Hence the first result.
z ∞ sin ap cos px RS
π, | x|< a UV
as
Again, from (i), 2 dp =

0 p T
0, | x|> a W
z ∞ sin ap cos px Rπ/2,
dp = S
| x| < a UV
g
⇒
0 p T 0, | x| > a W
.c
Putting x = 0 and a = 1, we get
z ∞ sin p π
w
dp = . ...(ii)
0 p 2
RS1 − x ,
w
2
if |x|< 1
Example 7. Find the Fourier transform of F(x) =
T 0, if |x|> 1
w
and use it to evaluate

x3 z FGH
0
∞ x x cos x − sin x
cos dx .
2
Sol. Fourier transform of F(x) is given by
IJ
K (U.P.T.U. 2015)
f(p) = z∞
–∞
eipx . F(x) dx = z
−1
1
eipx (1 – x2) dx
=
z1
−1
cos px (1 – x2) dx + i
z 1
−1
sin px (1 – x2) dx
=2 z 1
0
cos px (1 − x 2 ) dx | Using property of Definite Integral
www.cgaspirants.com
www.cgaspirants.com
LMR O
=2
N
2
Q
1
0
z
SMT(1 − x ) sinppx UVW − (− 2 x) . sinppx dxPP = 4p x sin px dx0
1
z 0
1
4 LR F − cos px I U F − cos px IJ dxOP = 4 LM − cos p + 1 FG sin px IJ OP

z
1 1
= M S xG
1
p MT H
J
KWV − 1. G
H p K PQ p MN p p H p K PQ
N p 0
0
0
4
= (sin p − p cos p)
p3
By inversion formula for Fourier transform
F( x) =
1
2π z ∞ 4
p3
(sin p − p cos p) . e − ipx dp
m
−∞
=
2
z∞
(cos px − i sin px) .
F sin p − p cos pI dp
GH JK
3
o
π −∞ p
4
z FGH sin p − p cos p I
.c
∞
=
π 0 p3
JK
. cos px dp | Using prop. of definite integral
Put x =
1
2
, F
FG 1IJ = 4
H 2K π
n z
0
∞ (sin p − p cos p)
p 3
p
. cos dp
2 ts
3
GH4
z
F p cos p − sin p I . cos p dp
JK 2
∞
ra
⇒ =− 3 ...(2)
4 π p 0
F 1I F 1I
F G J = 1− G J = 1− =
1 3
2
pi
where
H 2K H 2K 4 4
z
( p cos p − sin p)
∞ p 3π
as
From eqn. (2), 3

. cos dp = −
0 p 2 16
Replacing p by x, we get
g
z
( x cos x − sin x)
∞ x 3π
.c
3
cos dx = − .
0 x 2 16
Example 8. Find the Fourier transform of the function
w
F(t)
shown in the adjoining figure.
w
R|
2 , for − 1 < t < 1 U|
w
Sol. Here, S|
F(t) = 1, for − 2 < t < − 1 V| 1
T
1, for 1 < t < 2 W
–2 –1 0 1 2 t
Fourier transform is given by,
f(p) = z
−2
−1
1 . e ipt dt + z−1
1
2 . e ipt dt + z
1
2
1 . e ipt dt
Fe I +2Fe I +Fe I
ipt −1 ipt 1 ipt 2
=GH ip JK GH ip JK GH ip JK
−2 −1 1
=G
Fe − e I +2Fe − e I +Fe
− ip −2 ip ip − ip 2 ip
− e ip I
H ip JK GH ip JK GH ip JK
www.cgaspirants.com
www.cgaspirants.com
Fe ip
− e − ip I F
e 2ip − e −2ip I
= GH ip
+ JK GHip JK
1 1 2
= (2 sin p) + 2 sin 2 p = sin p (1 + 2 cos p).
p p p
Example 9. Find the Fourier sine transform of e–| x |. Hence evaluate z

0
∞ x sin mx
1 + x2
dx.
Sol. In the interval (0, ∞), x is positive so that e–| x | = e–x.

Fourier sine transform of f(x) = e–x is given by
Fs{f(x)} = z ∞
f(x) sin px dx = z ∞
e–x sin px dx
m
0 0
Le −x O ∞
p
=M (− sin px − p cos px)P =
o
N1 + p 2
Q 0
1 + p2
.c
Using inversion formula for Fourier sine transform, we get
f(x) =
2 ∞
z
F {f(x)} sin px dp or e–x =
π 0 s
2
π
ts z
0
∞ p
1 + p2
sin px dp
z z
n
2 ∞ p 2 ∞ x sin mx
Replacing x by m, we have e–m = 2
sin mp dp = dx
1+ p π 0 π 0 1 + x2
ra
Hence, z ∞
0
x sin mx
1 + x2
dx =
π
2
e–m.
pi
e − ax
Example 10. Find Fourier sine transform of , a > 0. Hence find Fourier sine
x
as
1
transform of . (G.B.T.U. 2011; U.P.T.U. 2008, 2015)
x
z z
g
∞ ∞ e − ax
Sol. fs (p) = F(x) sin px dx = . sin px dx = I (say) ...(1)
.c
0 0 x
dI d F
GH z ∞ e − ax I
JK
w
∴ = . sin px dx | Differentiating (1) w.r.t. p

dp dp 0 x
z z z
w
∞ e − ax ∂ ∞ e − ax ∞
= (sin px) dx = . x cos px dx = e–ax cos px dx
0 x ∂p 0 x 0
w
LM e (− a cos px + p sin px)OP

=
− ax ∞
=
a
Na + p 2
Q 2
0
a 2 + p2
Integration w.r.t. p yields,
F pI
I = tan GH JK + c –1 ...(2)
a
Initially when p = 0, I = 0 ∴ c=0 | From (2)
FG pIJ
∴ From (2), I = tan–1 H aK
∴ z0
∞ e − ax
x
. sin px dx = tan–1
FG pIJ
H aK
www.cgaspirants.com
www.cgaspirants.com
Take limit as a → 0
1
x z ∞
sin px dx =
0
π
2
Example 11. Find Fourier cosine transform of the following functions:
R| x, 0<x<
1
| 1
2
(i) F(x) = S 1 − x, < x<1 (U.P.T.U. 2009)
|| 0, 2
x>1
|T
(ii) F(x) = S
R|cos x, 0<x<a e ax + e − ax cosh ax
m
(iii) F(x) = or ,–π<a<π
|T 0, x>a e πx + e − πx cosh πx
o
x2
(iv) F(x) = e–2x + 4e–3x (v) F(x) = sin .
.c
2
Sol. (i) fc(p) = z ∞
F(x) cos px dx = z 1/2
x cos px dx + z 1
(1 – x) cos px dx
=G
H p K
0
F x sin px IJ 1/2
n
– z0
1/2
0
sin px
p
dx + (1 − x)
sin px
p
ts RS
T
1/2
UV
W
1
– z
1/2
1
(– 1)
sin px
p
dx
0 1/2
ra
1 p cos px F
I – 1 sin p – 1 FG cos pxIJ
1/2 1
=
2p
sin
2
+
p2
GH
JK 2 p 2 p H K 2
pi
0 1/2
1 FG cos p − 1IJ − 1 FG cos p − cos pIJ

=
H 2 K p H 2K
as
2
p2
1 FG 2 cos p − 1 − cos pIJ
=
p2 H 2 K
g
z
.c
∞
(ii) fc(p) = F(x) cos px dx
z z
0
w
a 1 a
= cos x cos px dx = [cos (1 + p) x + cos (1 – p)x] dx
0 2 0
w
=
LM
1 sin (1 + p) x sin (1 − p) x
+
OP a
=
LM
1 sin (1 + p) a sin (1 − p) a
+
OP
2 1+ p N 1− p Q 2 N
1+ p 1− p Q
w
I FG e I
(iii) fc(p) = z
0
∞ Fe
GH e
ax
πx
+ e − ax
+e − πx
I cos px dx = F e
JK GH e z ∞
0
ax
πx
+ e − ax
+e − πx JK H
ipx
+ e − ipx
2 JK dx
=
1
2
LM e
N z
e +e
0
+e
∞ ( a + ip) x
dx + πx
e
e +e
− ( a + ip) x
+e
− πx z ∞
0
( a − ip) x
πx
− ( a − ip) x
− πx
dxP
O
Q
L OP
1 L1 F a + ip I 1 F a − ip I O 1M
H 2 JK + 2 sec GH 2 JK PQ = 4 MM cos FG a + ipIJ + cos FG a − ipIJ PP
1 1
sec G
2 MN 2
=
MN H 2 K H 2 K PQ
www.cgaspirants.com
www.cgaspirants.com
a ip a p
2 cos
cos cos cosh
1 2 2 2 2
= . = ∵ cos ip = cosh p
4 FG
a + ip a − ipIJ FG IJ cos a + cosh p
cos
2 H cos
2 K H K
(iv) fc(p) = z0
∞
(e–2x + 4e–3x) cos px dx
= z0
∞
e–2x cos px dx + 4 z ∞
0
e–3x cos px dx = 2
p +4
2
+ 2
p +9
12
R|sin F x + pxI + sin F x I U|V dx

z ∞ x2 1
z ∞
S| GH 2 JK GH 2
2 2
JK W|
m
(v) fc (p) = sin cos px dx = − px
0 2 2 0
T
z Fx I
z F x + pxI dx = 1
z F x + pxI dx
o
2
1 ∞ 2 1 –∞ 2 ∞
sin G
= sin GH 2 + px dx – JK sin G
H 2 JK 2 H 2 JK
.c
2 0 2 0 −∞
Fx I
z ts z
2
1 ∞ i GH 2 + px JK 1 ∞ i ( x 2 + 2 px )
= Im. part of e dx = Im. part of e2 dx
2 −∞ 2 −∞
n
1
z ∞ 1 − ip2 /2
z ∞ −
( x + p) 2
ra
i {( x + p) 2 − p2 } 2i
= Im. part of e 2
dx = Im. part of e e dx
2 −∞ 2 −∞
1 −(ip2 / 2)
pi
π
= Im. part of e
2 1
as
2i
π F cos p − i sin p I FG cos π + i sin π IJ

2 2
= Im. part of GH 2 2 JK H 4K
g
2 4
.c
π F cos p 2
p I 1 π F 2
p p I 2 2
= GH 2 − sin J
2 K
. = G cos
H 2 − sin
2 JK
.
w
2 2 2
1
w
Example 12. Find Fourier cosine transform of and hence find Fourier sine
1 + x2
w
x
transform of . (M.T.U. 2014)
1 + x2
Sol. fc(p) = z
0
∞
1
1 + x2
cos px dx = I (say) ...(1)
∴
dI
dp
= z
∞
0
− x sin px
1 + x2
dx = – z 0
∞ (1 + x 2 − 1) sin px
x(1 + x 2 )
dx
=– z
0
∞ sin px
x
dx + z0
∞ sin px
x(1 + x 2 )
dx
dI
dp
=–
π
2
+ z ∞
0
sin px
x(1 + x 2 )
dx ...(2)
www.cgaspirants.com
www.cgaspirants.com
Again,
d 2I
dp2
= z 0
∞ x cos px
x (1 + x ) 2
dx = z0
∞ cos px
1 + x2
dx = I | From (1)
d 2I
⇒ –I=0 ...(3)
dp2
Solution of (3) is, I = c1 ep + c2 e–p ...(4)
dI
∴ = c1 ep – c2 e–p ...(5) | From (4)
dp
When p = 0,
dI
I=
∞ 1
0 1+ x
π
z2 dx =
π
2
| From (1)
m
and =– | From (2)
dp 2
Applying to (4) and (5), we get
o
c1 + c2 = π/2 and c1 – c2 = – π/2
.c
so that, c1 = 0, c2 = π/2
π –p
From (4), eI=
ts
∴
2
z
∞ cos px π –p
n
⇒ dx = e
0 1+ x 2 2
ra
Differentiating w.r.t. p, we get
z ∞ − x sin px
dx = –
π –p
e ⇒ z ∞ x sin px
dx =
π –p
e .
pi
2
0 1+ x 2 0 1+ x 2 2
2
Example 13. Find the Fourier cosine transform of e − x .
as
2
Sol. Fourier cosine transform of e − x is given by
z ∞
g
2 2
Fc { e − x } = e − x cos px dx = I (say) ...(1)
0
.c
Differentiating w.r.t. p, we have

dI
zL 1 ∞
z
w
∞ 2 2
=– x e − x sin px dx = (sin px) (– 2x e − x ) dx
dp 0 2 0
z OP
w
1 2 ∞ ∞
MN{ }
2
= sin px e − x −p cos px e − x dx (Integrating by parts)
2 0 0 Q
w
dI
p ∞ − x2
=–
2 0
p
e z
cos px dx = –
p
2
I
=– dp
dp 2
p2
p2 −
4
Integrating, we have log I = – + log A or I=A e ...(2)
4
Now when p = 0, from (1), I = z

0
∞ 2
e − x dx =
2
π
π
∴ From (2), =A
2
www.cgaspirants.com
www.cgaspirants.com
p2
− x2 π − 4
Hence, I = Fc { e }= e .
2
1 F I 2
Example 14. If f c (p) =
2 GH JK
tan −1 2 , then find F(x).
p
(M.T.U. 2014)
F 2 I = tan LM 2 OP = tan F 1 I − tan F 1 I

Sol. tan −1 GH p JK
2
−1
N ( p − 1) + 1Q 2 GH p − 1JK GH p + 1JK −1 −1
F( x) =
2
π 2
1
z∞
tan G
0
F 2 I cos px dp
H p JK
−1
2
1 R
z
STtan FGH p 1− 1IJK − tan FGH p 1+ 1IJK UVW cos px dp
∞
−1 −1
m
=
π 0
1L
z
F 1 IJ cos px dp − tan FG 1 IJ cos px dpOP
z ∞
o
∞
= M tan G −1 −1
πN H p − 1K
0 H p + 1K Q 0
.c
= I1 + I2 (say) ...(1)
LMR F 1 I sin px U OP
where I1 =
1
π MNSTtan GH p − 1JK . x VW
−1
n ∞
0
− z
0
∞
ts
−1
( p − 1) 2
.
1
RS1 + 1 UV .
sin px
x
dp
Q
T ( p − 1) W 2
z
ra
1 ∞ sin px
= dp
πx 0 ( p − 1) 2 + 1
z
pi
−1 ∞ sin px
Similarly, I2 = dp
πx 0 ( p + 1) 2 + 1
as
From (1), I=
1
πx z RST
0
∞ sin px
( p − 1) 2 + 1
−
sin px
( p + 1) 2 + 1
UV dp
W
g
1 π −x π
e sin x − e − x
=
LM O
sin (− x)P = e
−x
sin x
.c
2πx 1 1 N Q x
.
Example 15. Solve the integral equations:
w
R| 1,
z z
0≤t<1
∞ ∞
w
(i) f(x) cos λx dx = e −λ

(ii) F(x) sin tx dx = S 2, 1 ≤ t < 2.
0 0
|T 0, t≥2
w
Sol. (i) f ( x) =
2
π z0
∞
e – λ cos λx dλ =
2
.
1
π 1 + x2
∵ f c (λ ) = e – λ
(ii) F( x) =
2
π z∞
0
fs (t) sin tx dt =
2
π
LM
N z
0
1
sin tx dt + z 2
1
2 sin tx dt
OP
Q
LMFG − cos tx IJ − 2 FG cos tx IJ OP
2
1 2
MNH x K H x K PQ
=
π 0 1
2 L 1 cos x F cos 2x − cos x IJ OP = 2 (1 + cos x − 2 cos 2x) .

= M − −2G
π Nx x H x x K Q πx
www.cgaspirants.com
www.cgaspirants.com
Example 16. Solve the integral equation
z 0
∞
F(x) cos px dx =
RS1 − p,
T 0,
0≤ p≤1
p>1
Hence prove that z 0

∞ sin 2 t
t 2
dt =
π
2
.
Sol. Let z0
∞
F( x) cos px dx = fc ( p) , then fc ( p) =
RS1 − p,
T 0,
0≤ p≤1
p>1
∴ By inversion formula for Fourier cosine transform, we have
F( x) =
2
z ∞
fc ( p) cos px dp =
2 LM
N z 1
(1 − p) cos px dp + z ∞
0 . cos px dp
OP
Q
m
π 0 π 0 1
LM OP 1
LM OP
o
2 sin px − cos px 2 cos x 1 2(1 − cos x)
= (1 − p) . − (– 1) . = − + 2 =
π N x x2 Q π N x2 x Q
πx 2
.
.c
0
Deduction. Since
z 0
∞
F( x) cos px dx =
n RS1 − p,
T 0,
0≤ p≤1
ts
p>1
where F(x) =
2(1 − cos x)
πx 2
∴
2
z ∞ 1 − cos x
. cos px dx =
1 − p, 0 ≤ p ≤ 1 RS
ra
π 0 x 2
0, p > 1 T
x
z z 2 sin 2
pi
2 ∞ 1 − cos x ∞
2 dx = π
When p = 0, we have dx = 1 or
π 0 x2 0 x2 2
z
as
∞ sin 2 t π
Putting x = 2t so that dx = 2dt, we get dt = .
0 t2 2
RS1,
g
0<x<π
Example 17. Taking the function F(x) =
T0,
.c
x>π
FG 1 − cos pπ IJ . sin px dp = R|S π2 ,

z
w
∞ 0<x<π
show that .
H p K
0
|T 0, x>π
w
Sol. Consider F( x) = S
R1, 0 < x < π
T0 , x > π
w
Taking sine transform,
fs ( p) = z
0
∞
F( x) . sin px dx =
z π
0
1 . sin px dx =
1 − cos pπ
p
By inversion formula for Fourier sine transform.
2
π z FGH
0
∞
p
IJ
1 − cos pπ
K sin px dp =
RS
1, 0 < x < π
T
0, x > π
R| π , 0 < x < π
⇒ z FGH
0
∞ 1 − cos pπ I
p
JK sin px dp = S| 2
T 0, x > π.
www.cgaspirants.com
www.cgaspirants.com
Example 18. Find Fourier sine and cosine transform of F(x) = xn e–ax ; a > 0, n > – 1
Hence find Fourier sine and cosine transforms of
(i) xm–1 (ii) x–m.
Sol. Here F(x) = xn e–ax
fs(p) = z0
∞
xn e–ax . sin px dx = z 0
∞
xn e–ax .
Fe
GH
ipx
− e − ipx
2i
I dx
JK
=
1
2i z
LM x e
N0
∞
n − ( a − ip) x
dx − z 0
∞
x n e − ( a + ip) x dx
OP
Q
=
1 LM Γ(n + 1)
N (a − ip)
−
Γ (n + 1) OP
Q
∵ z ∞
e − zx x n − 1 dx =
Γ(n)
m
2i n+ 1
(a + ip) n + 1 0 zn
Put a = r cos θ, p = r sin θ
o
so that, a + ip = r(cos θ + i sin θ)
.c
∴ (a + ip)n+1 = rn+1 {cos (n + 1) θ + i sin (n + 1) θ}
and a – ip = r (cos θ – i sin θ)
∴
Also,
(a – ip)n+1 = rn+1 {cos (n + 1) θ – i sin (n + 1) θ}
r2 = a2 + p2 and
n
tan θ =
ts p
a
ra
Γ(n + 1) LM 2i r n+1
sin (n + 1) θ OP
Now, fs(p) =
2i N r 2(n + 1) Q
pi
Γ(n + 1) sin (n + 1) θ Γ(n + 1) sin (n + 1)θ p

= n+1 = where tan θ =
r 2
(a + p ) 2 ( n + 1)/ 2 a
as
Γ(n + 1) cos (n + 1)θ p

Similarly, fc (p) = 2 2 ( n + 1)/ 2 where tan θ =
(a + p ) a
g
∴ We have the relations,

.c
z ∞
e–ax . xn sin px dx =
Γ(n + 1) sin (n + 1)θ
...(1)
w
0 (a 2 + p2 ) ( n + 1)/ 2
z ∞ Γ(n + 1) cos (n + 1)θ

w
and e–ax xn cos px dx = ...(2)

0 (a 2 + p2 ) ( n + 1)/ 2
w
p
where tan θ =
a
Put a = 0 and replace n by m – 1 in (1) and (2), we get
z
∞
0
xm–1 sin px dx =
pm
Γ(m) sin m π / 2
...(3)
and
∞
0 z
xm–1 cos px dx =
Γ(m) cos m π / 2
pm
Replace m by m + 1 in (3) and (4), we get
...(4)
z
0
∞
xm sin px dx =
Γ(m + 1) sin (m + 1) π/2
p m+ 1 =
Γ(m + 1) cos m π / 2
pm +1
...(5)
www.cgaspirants.com
www.cgaspirants.com
and
0 z ∞ − Γ(m + 1) sin m π / 2
xm cos px dx =
pm+ 1
Replacing m by – m in (5) and (6), we get
...(6)
0 z
x–m sin px dx =
Γ(1 − m) cos m π / 2
p1 − m
π
π 1 cos m π / 2 ∵ Γ (m) Γ (1 − m) = ;
= . . sin mπ
sin m π Γ(m ) p1−m 0< m<1
π
π cos m
p m−1 2 π p m−1 mπ
= . = . cosec
m
Γ(m) 2 sin m π cos m π 2 Γ( m ) 2
2 2
o
∞ π p m−1 mπ
Similarly, x–m cos px dx = . sec
2 Γ( m ) 2
.c
0
n ASSIGNMENT
R| 0, for x < a U|
ts
1. Find Fourier transform of F(x) = S| 1, for a < x < b . V|
ra
T 0, for x > b W
2. (i) Show that the Fourier transform of
pi
F(x) =
RSa −| x|,
T 0,
for | x | < a
for | x | > a > 0
2 UV
is 2 (1 – cos ap). Hence show that
p W z0
∞ sin2 t
t2
dt =
π
2
.
as
(ii) Find the Fourier transform of
F(x) =
| x|< 1RS1, UV
hence evaluate
∞ sin x
dx. z (G.B.T.U. 2011)
g
| x|> 1 T0, W 0 x
.c
3. If f(p) is the Fourier transform of F(x), prove that F[eiax F(x)] = f(p + a). F(t)
1
w
4. Find the Fourier transform of the single gate function (rectangular pulse)
w
shown in the adjoining figure.

– T/2 T/2 t
w
F(t)
5. Find the Fourier transform of the function F(t)as shown in the figure:
LM U|OP
R| At , 0<t<T
MNHint. F(t) = S|A,T
V
T < t < 2T |WPQ
A
M N
T
Find Fourier transform of F(x) = S0,
Rx , 2
| x |< a UV
6.
T | x|> a . W O T 2T t
7. Find Fourier sine and cosine transform of
Fig. of Q. 5
(i) F(x) = e–x, x ≥ 0 (ii) F(x) = 2e–5x + 5e–2x
(iii) F(x) = cosh x – sinh x (iv) F(x) =

RS1, 0≤ x<a UV
T0 , x>a W
(v) x e–ax, a > 0.
www.cgaspirants.com
www.cgaspirants.com
x, R| for 0<x<1 U|
8. Obtain Fourier cosine transform of F(x) = 2 − x, S| for 1< x < 2 V|
0, T for x>2 W
Also find its Fourier sine transform.
9. Find Fourier sine transform of
R|
0, 0 < x < a U| 1
S|
(i) F(x) = x, a ≤ x ≤ b V| (ii) F(x) =
x ( x 2 + a2 )
T
0, x > b W
2 eax + e− ax cosh ax
(iii) F(x) = x e− x /2
(iv) F(x) = or ,–π<a<π
eπx − e− πx sinh πx
m
e − ax − e −bx
(v)
x
o
1 1
10. Find Fourier sine transform of and deduce that fs (cosec h πx) = tanh (p/2)
.c
e −e πx − πx 2
11. Find Fourier cosine transform of F(x) = sech πx.
12.
13.
Find Fs–1
Fe I
GH p JK and hence evaluate F FGH 1p IJK .
− ap
By taking e–ax for F(x); a > 0, x > 0; show that

n s
–1
ts
z z
ra
∞ cos px ∞ p sin px
π –ax π –ax
2 e 2 dp =
and dp = e .
0 a + p 2a 0 a2 + p2 2
pi
14. Find F(x) if its Fourier sine transform is
(i)
p
(ii)
π
(iii) (2πp)1/2 (iv)
RSsin p, 0< p< π
.
UV
1 + p2 T 0, W
as
2 p≥π
2 2 2 2
15. Find Fourier cosine transform of e− a x
and hence evaluate Fourier sine transform of x e− a x
.
1
g
16. Find Fourier sine and cosine transform of .

x
.c
dn
17. If f(p) is the Fourier transform of F(x), prove that F[xn F(x)] = (– i)n {f(p)}.
w
dpn
1
w
2
/( p2 + 1)}
18. A certain function of time F(t) has the following Fourier transform f(p) = e− { 2 p
p2 + 1
w
Using the properties of the Fourier transform, obtain the Fourier transforms of
(i) F(2t) (ii) F(t – 2)
(iii) F(t) cos 2t (iv) e2it F(t).
19. State and prove the convolution theorem for the Fourier transform. Verify this theorem for the
functions f(t) = e–t and g(t) = sin t. (U.P.T.U. 2010)
Answers
( eibp − eiap ) 2 sin p π
1. 2. (ii) ,p≠0;
ip p 2
FG pT IJ FG where sin c θ = sin θ IJ A
( eipT − 1) +
A 2ipT
e
4. T sin c H 2 KH θ K
5.
Tp 2 ip
www.cgaspirants.com
www.cgaspirants.com
F 2a 2
4 I
6. GH p − J sin ap + 4pa cos ap
p K
3 2
p
;
1 F 2 5 I ; 10 F 1 1 I
7. (i)
1 + p2 1 + p2
(ii) p GH p 2
+ 25
+
p 2 J GH p
+ 4K 2
+ 25
+ 2
p
J
+ 4K
p
;
1 FG 1 − cos ap IJ ; sin ap 2ap
;
a2 − p2
(iii)
1+ p 2
1+ p 2 (iv)
H p K p (v) 2
(a + p ) 2 2
(a2 + p2 )2
2 2
8. 2 cos p (1 – cos p) ; sin p (1 – cos p)
p p2
( a cos ap − b cos bp) (sin bp − sin ap) π
m
9. (i) + (ii) (1 – e–ap)
p p2 2a2
π sinh p FG p IJ − tan FG p IJ −1
o
2
(iii)
2
p e−( p / 2) (iv)
2 (cos a + cosh p)
(v) tan–1
H aK H bK
.c
10.
1
tanh
p
11.
1
sech
p
12.
2 F xI
tan GH JK ; 1 –1
4 2 2 2 π a
14. (i) e–x
π − ( p2 / 4 a 2 ) p π − ( p2 / 4 a 2 )
(ii)
1
x
n
π
(iii)
π
x x
1
ts 2 sin πx
(iv) π .
(1 − x 2 )
15. e ; e 16. ,
ra
2a 4 a3 2p 2p
F 2p I 2
2
− GH p + 4 JK 1
pi
2
2
/ ( p2 + 1)}
18. (i) e (ii) e2ip . 2
e− {2 p
p2 + 4 p +1
LM |RS 2( p + 2) |UV |RS 2( p − 2) |UV O

as
2 2
|RS 2( p + 2) |UV
2
T| ( p − 2) + 1 W| PP
− −
(iii)
1
MM ( p + 21) e T| ( p + 2) + 1 W| +
2 1
e
2
(iv)
1
e
−
T| ( p + 2) + 1 W| .
2
2
MN
2
+1 ( p − 2)2 + 1 PPQ ( p + 2)2 + 1
g
.c
2.11 FOURIER TRANSFORMS OF THE DERIVATIVES OF A FUNCTION

w
Let u ( p, t ) be the Fourier transform of the function u(x, t).
z
w
∞
Then u ( p, t) = u( x, t) e ipx dx
−∞
w
∂u ∂ 2u
Suppose u and both vanish as x → ± ∞. Then the Fourier transform of is given
∂x ∂x 2
by
F
RS ∂ u UV =
2
T ∂x W 2 z −∞
∞ ∂ 2u ipx
∂x 2
e dx
LMRe OP
=
MNST
ipx ∂u
∂x
− ipe ipx . u
UV
W
∞
−∞
+ z−∞
∞
(ip) 2 e ipx . u dx = − p2
PQ z
−∞
∞
ue ipx dx = − p2 u
F ∂ uI 2
Hence, F GH ∂x JK = – p
2
2
u where u = F(u)
www.cgaspirants.com
www.cgaspirants.com
If us ( p, t) and uc ( p, t) be the Fourier sine and cosine transforms of u(x, t), then
us ( p, t) = z
0
∞
u sin px dx and
∂ 2u
u ( p, t) = z ∞
0
u cos px dx
The Fourier sine transform of is given by

∂x 2
Fs
RS ∂ u UV =
2
T ∂x W2 z
0
∞ ∂ 2u
∂x 2
sin px dx
LMRsin px ∂u − p cos px . uU − p sin px . u dxOP

MNST
=
∂x
VW PQ
∞
0
z0
∞
2
L R U − p u sin px dxOP R U
m
= M pSuV
MN T W
2
z
0
∞
PQ = pSTuVW − p u
2
s
o
x=0 x=0
F ∂ uI
.c
2
F G
H ∂x JK = p(u) – p u
Hence, 2
where u = F (u)
s 2 x=0 s s s
Similarly, we have,
n ts
F ∂ u I = – FG ∂uIJ
2
ra
Fc GH ∂x JK H ∂x K
2
x=0
– p2 u c where uc = Fc(u) (U.P.T.U. 2008)
pi
In general, the Fourier transform of the nth derivative of F(x) is given by
F d FI n
as
F GH dx JK = (– ip)
n
n F{F(x)}
g
Note. s may also be used as a parameter in place of p.

.c
2.12 CHOICE OF INFINITE FOURIER SINE OR COSINE TRANSFORM

w
∂ 2u
For exclusion of from a differential equation, we require
∂x 2
w
∂u FG IJ
(i) (u)x = 0 in sine transform. (ii) ∂x in cosine transform. H K
w
x=0
2.13 APPLICATIONS OF FOURIER TRANSFORMS TO HEAT CONDUCTION

(TRANSFER) EQUATIONS
In one dimensional heat transfer eqns, the partial differential equation can easily be trans-
formed into an ordinary differential equation by applying Fourier transforms. The required
solution is then obtained by solving this equation and inverting by means of the complex inversion
formula. This is illustrated through the following examples.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
∂u ∂ 2 u
Example 1. Solve the equation = , x > 0 , t > 0 subject to the conditions
∂ t ∂x 2
(i) u = 0 when x = 0, t > 0 (ii) u =

RS
1, 0 < x < 1
when t = 0
0, x ≥ 1 T
and (iii) u(x, t) is bounded. (G.B.T.U. 2011; U.P.T.U. 2009, 2015)
Sol. Since (u)x = 0 is given, taking Fourier sine transform of both sides of the given
equation, we have
z ∞ ∂u
z ∞ ∂ 2u
m
sin px dx = sin px dx
0 ∂t 0 ∂x 2
Hz
d FG ∞ IJ
o
u sin px dx = p(u)
⇒
dt 0 K2
x=0– p u
z
.c
du ∞
⇒ + p2 u = 0 ...(1) where u = u sin px dx
dt 0
Solution to (1) is u = c1 e − p t ,
where c1 is a constant.
2
n ts ...(2)
Now, when t = 0, Fourier sine transform of u(x, t)

ra
( u )t = 0 = z ∞
u(x, 0) sin px dx = z 1
1. sin px dx + z ∞
0 . sin px dx
pi
0 0 1
F − cos px IJ
=G
1
=
1 − cos p
H p K p
as
0
∴ From (2), ( u ) t = 0 = c1
1 − cos p
g
⇒ c1 =
p
.c
∴ From (2), u =
FG 1 − cos pIJ e− p
2
t
H p K
w
Applying inverse Fourier sine transform, we have
z FGH
w
u(x, t) =
2 ∞
e
IJ
1 − cos p − p2t
sin px dp
π p K
w
which is the required solution.

Example 2. Determine the distribution of temperature in the semi-infinite medium x ≥ 0
when the end x = 0 is maintained at zero temperature and the initial distribution of tempera-
ture is F(x).
Sol. Let u(x, t) be the temperature at point x at any time t. Heat flow equation is
∂u ∂ 2u
= c2 , (x > 0, t > 0) ...(1)
∂t ∂x 2
Subjected to the initial condition u(x, 0) = F(x) ...(2)
and the boundary condition u(0, t) = 0 ...(3)
www.cgaspirants.com
www.cgaspirants.com
Taking Fourier sine transform of eqn. (1), we get
z0
∞ ∂u
∂t
sin px dx = c2 z ∞
0
∂ 2u
∂x 2
sin px dx
du
⇒ = c2 [p (u)x = 0 – p2 u ] = – c2p2 u ∵ (u) x = 0 = 0
dt
⇒
du
dt
+ c 2 p2 u = 0
2 2
...(4) where u = z
0
∞
u sin px dx
−c p t
Solution to (4) is u = c1 e ...(5)
Taking Fourier sine transform of (2), we get
m
∞
( u )t = 0 = F( x) sin px dx = fs(p) | say
0
From (5), ( u ) t = 0 = c1 ⇒ c1 = fs(p)
o
2 2
−c p t
∴ From (5), u = fs(p) e
.c
Now Taking its inverse Fourier sine transform, we get
u(x, t) =
2
π
n z ∞
0
fs ( p) e − c
2
p2 t
ts
sin px dp
∂u ∂ 2u
Example 3. Use Fourier sine transform to solve the equation =2 under the
∂x 2
ra
∂t
conditions
(i) u(0, t) = 0 (ii) u(x, 0) = e–x (iii) u(x, t) is bounded.
pi
Sol. The given equation is

∂u ∂ 2u
=2 ...(1)
as
∂t ∂x 2
Taking Fourier sine transform on both sides of eqn. (1), we get
z z
g
∞ ∂u ∞ ∂ 2u
sin px dx = 2 sin px dx
.c
0 ∂t 0 ∂x 2
du
⇒ = 2 [p(u)x = 0 – p2 u ]
w
dt
⇒
du
+ 2p2 u = 0 where u = z ∞
u sin px dx
w
dt 0
2
Its solution is u = c1 e − 2 p t
...(2)
w
where c1 is a constant.
At t = 0,
From (2),
( u )t = 0 = z
0
( u ) t = 0 = c1
∞
(u) t = 0 sin px dx = z0
∞
e − x sin px dx =
p
1 + p2
...(3)
...(4)
p
∴ From (3) and (4), c1 =
1 + p2
p 2
From (2), u = 2
e− 2 p t
...(5)
1+ p
Taking inverse Fourier sine transform, we get
u(x, t) =
2
π z ∞
0 1+ p
p
2
2
e −2 p t sin px dp.
www.cgaspirants.com
www.cgaspirants.com
Example 4. The temperature u in the semi-infinite rod 0 ≤ x < ∞ is determined by the

∂u ∂ 2u
differential equation =k subject to conditions
∂t ∂x 2
∂u
(i) u = 0 when t = 0, x ≥ 0 (ii) = – μ (a constant) when x = 0 and t > 0
∂x
Making use of cosine transform, show that u(x, t) =

2μ
π z
0
∞ cos px
p 2
2
(1 − e − kp t ) dp.
Sol. Taking Fourier cosine transform on both sides of given equation, we get
z ∞ ∂u
. cos px dx = k zL ∞ ∂ 2u
cos px dx
m
0 ∂t 0 ∂x 2
⇒
du
=k −
∂u
MN FGH IJK − p2u
OP
Q
o
dt ∂x x =0
z
.c
∞
= kμ – kp2 u where u = u cos px dx
0
⇒
du
dt
+ kp2 u = kμ
I.F. = e
n
kp t 2
ts ...(1)
ra
Solution to (1) is
u . e kp2 t
= z 2
kμ . e kp t dt + c1 =
μ
p 2
e kp2 t
+ c1
pi
μ 2
⇒ u = 2
+ c1 e − kp t
...(2)
p
as
At t = 0, ( u )t = 0 = z ∞
0
(u) t = 0 cos px dx = 0 ...(3)
g
μ
From (2), ( u )t = 0 =
+ c1 ...(4)
.c
p2
μ
⇒ c1 = – 2 | From (3) and (4)
w
p
μ − kp2 t
∴ From (2), u = 2 (1 – e )
w
p
Taking inverse Fourier cosine transform, we get
w
u=
2 ∞ μ
π 0 p2
2
z
(1 − e – kp t ) cos px dp.
Example 5. (i) If the initial temperature of an infinite bar is given by
θ(x) =
RSθ , for| x|< a
0
T0, for| x|> a
determine the temperature at any point x and at any instant t.
(ii) If the initial temperature of an infinite bar is given by
RS1,
μ(x, 0) = 0, otherwise
for − c < x < c UV
T W
determine the temperature of an infinite bar at any point x and at any time t > 0.
www.cgaspirants.com
www.cgaspirants.com
Sol. (i) To determine the temperature θ(x, t), we have to solve the heat-flow equation
∂θ ∂ 2θ
= c2 2 , t>0 ...(1)
∂t ∂x
subject to the initial condition θ( x, 0) =

RSθ , for|x|< a
0 ...(2)
T0, for|x|> a
Taking Fourier transform of (1), we get
z ∞∂θ ipx
−∞ ∂t
e dx = c 2 z−∞
∞ ∂ 2θ
∂x 2
e ipx dx
or
d
dt z ∞
θ e ipx dx = − c 2 p2 θ
m
−∞
or
dθ
dt
= − c 2 p2 θ ...(3) where θ = z ∞
θ e ipx dx
o
–∞
Now taking the Fourier transform of (2), we get
.c
z z L e OP ipx a
∞ a
θ ( p, 0) = θ( x, 0) e ipx
dx = θ e dx = θ M
ipx
N ip Q
ts
0 0
−∞ −a
−a
LM e ipa
−e − ipa OP = 2θ LM e − e OP = 2θ sin pa
ipa − ipa
n
0 0
= θ0 ...(4)
N ip Q p N 2i Q p
ra
dθ
From (3), = − c 2 p2 dt
θ
pi
2
p2 t
Integrating, log θ = − c 2 p2 t + log A or θ = A e−c
2θ 0 sin pa
as
2θ 0 sin pa
Since θ = when t = 0, from (4), we get A=
p p
g
2θ 0 sin pa − c2 p2 t
∴ e θ=
p
.c
Taking its inverse Fourier transform, we get
z
w
1 ∞ 2θ 0 sin ap − c2 p2t − ipx

θ( x, t) = .e .e dp
2π −∞ p
w
=
θ0
z ∞ sin ap − c2 p2 t
.e (cos xp − i sin xp) dp
w
π −∞ p
Nz z
=
θ0 LM ∞ sin ap − c2 p2t
.e cos xp dp − i
∞ sin ap − c2 p2 t
.e sin xp dp
OP
π −∞ p −∞ p Q
=
θ0
π z −∞
∞sin ap − c 2 p2t
p
.e cos xp dp
(The second integral vanishes since its integrand is an odd function)
=
2θ 0
π z ∞
0
sin ap − c 2 p2 t
p
.e
θ
cos xp dp = 0
π z 0
∞
p
2
e−c p t
2
. 2 sin ap cos xp dp
=
θ0
π z ∞
0
e−c
2
p2 t FG sin (a + x) p + sin (a − x) pIJ dp
H p K
www.cgaspirants.com
www.cgaspirants.com
=
θ0
π z
0
∞ 2
e − v sin
RS
T
(a + x)v
c t
+ sin
(a – x)v dv
c t v
UV
where v2 = c2p2t
W
=
θ0
2
erf
RS F
a+x
T GH
2c t
+ erf
a− x
2c t
IJ ∞
−t 2 FG dt π IJ UV
. ∵ 0 e sin ( ct / x ) . t = 2 erf
K H KW
c
2 x
. z FG
H
IJ
K
(ii) To determine the temperature μ(x, t), we have to solve the heat-flow equation
∂μ ∂ 2μ
= k2 ,t>0 ...(1)
∂t ∂x 2
subject to the initial condition μ(x, 0) = 1, for − c < x < c ...(2) UV
0, otherwise W
Taking Fourier transform of (1), we get
m
∞ ∂μ ipx
− ∞ ∂t
. e dx = k2
∞ ∂2μ
− ∞ ∂x 2
. e ipx dx z
z
o
d ∞
⇒ μ e ipx dx = – k2p2 μ
.c
dt − ∞
⇒
dμ
dt
= – k2 p2 μ
n ts where μ = z
−∞
∞
μ e ipx dx
dμ
⇒ = – k2p2 dt
μ
ra
Integrating, we get
log μ = – k2p2t + log A | A is a constant
pi
2 2
or, μ = A e −k p t ...(3)
Now, taking Fourier transform of (2), we get
z
as
∞
μ (p, 0) = μ(x, 0) eipx dx
−∞
z
g
c
e ipc − e − ipc 2
= eipx dx = = sin cp ...(4)
.c
−c ip p
2
w
From (3), At t = 0, sin cp = A

p
w
2 2 2
∴ μ = sin cp e − k p t
p
w
Taking its inverse Fourier transform, we get
μ(x, t) =
which is the required solution.
1 ∞ 2
2π − ∞ p
2 2
z
sin cp e − k p t . e–ipx dp
Example 6. Use Fourier cosine transform to show that the steady temperature u in the
semi-infinite solid y > 0 when the temperature on the surface y = 0 is kept at unity over the strip
| x | < a and at zero outside the strip is
1 LM
tan −1
a+ x FG
+ tan −1
a−x IJ FG IJ OP
π MN y H y K H K PQ
The results z
0
∞
e − px x −1 sin rx dx = tan −1
FG r IJ (r , p > 0) may be assumed.
H pK
www.cgaspirants.com
www.cgaspirants.com
∂ 2u ∂ 2u
Sol. Taking Fourier cosine transform of + = 0 , we have
∂x 2 ∂y 2
z0
∞ ∂ 2u
∂x 2
. cos px dx + z ∞
0
∂ 2u
∂y 2
. cos px dx = 0
FG ∂uIJ
–
H ∂x K x=0
− p2 u +
d2
dy 2
(u ) = 0 where u = z0
∞
u cos px dx
d2 u ∂u
⇒ − p2 u = 0 ∵ → 0 as x → ∞ ...(1)
dy 2 ∂x
Its solution is u = c1e py + c2 e − py ...(2)
m
But u is finite so c1 = 0, otherwise u → ∞ as y → ∞
∴ From (2), u = c2 e − py ...(3)
o
z ∞
.c
But u= u cos px dx
z
0
From (3),
(u ) y = 0 =
(u ) y = 0 = c2
z
0
∞
n (u) y = 0 cos px dx =
ts
0
a
1 . cos px dx =
sin pa
p
...(4)
sin pa
ra
∴ c2 =
p
sin pa − py
u= e
pi
∴ From (3),
p
Applying inverse Fourier cosine transform, we get
z z
as
2 ∞ sin pa − py 1 ∞ e − py
u= e cos px dp = (2 sin pa cos px) dp
π 0 p π 0 p
z
g
− py
1 ∞ e
= [sin (a + x) p + sin (a − x) p] dp
.c
π 0 p
1 LM a+ x FGa−x IJ FG IJ OP .
w
= tan −1 + tan −1
π N y H y K H KQ
w
ASSIGNMENT
w
1. Apply appropriate Fourier transform to solve the partial differential equation

∂V ∂2 V
= ; x > 0, t > 0
∂t ∂x2
subject to the conditions
RS
x, 0 ≤ x ≤ 1
(i) Vx (0, t) = 0
T
(ii) V(x, 0) = 0, x > 1 (iii) V(x, t) is bounded.
∂u ∂ 2u
2. Solve = k 2 for x ≥ 0, t ≥ 0 under the conditions
∂t ∂x
(i) u(0, t) = u0, t > 0 (ii) u(x, 0) = 0, x ≥ 0 (iii) u(x, t) is bounded.
www.cgaspirants.com
www.cgaspirants.com
3. Using the method of Fourier transform, determine the displacement y(x, t) of an infinite string
given that the string is initially at rest and that the initial displacement is f(x), – ∞ < x < ∞. Show
1
that the solution can also be put in the form y(x, t) = [ f ( x + ct) + f ( x − ct)] .
2
4. Using Fourier transform, solve
∂V ∂2 V
= , − ∞ < x < ∞, t > 0 ; V ( x, 0) = f ( x) (U.P.T.U. 2008, 2014)
∂t ∂x2
5. Using Fourier transforms, solve the following initial boundary value problem:
∂u 2
2 ∂ u
1, − 1 < x < 0 U|
∂t
=c
∂x 2
, − ∞ < x < ∞, t > 0 ; u( x, 0) = −1, 0 < x < 1 V| (U.P.T.U. 2008)
W
m
0, otherwise
Answers
z FGH F I
o
2 ∞ sin p cos p − 1 − p2 t
1. V(x, t) = + e JK cos px dp
.c
π 0 p p2
2. u(x, t) =
2u0
π z GGH
0
∞ 1 − e− kp
p
2
t I
JJ sin px dp.
K ts
z z
n
1 ∞ 2 ∞
4. V(x, t) = F( p) e− p t . e− ipx dp where F( p) = f ( x) eipx dx
2π −∞ −∞
ra
5. u(x, t) =
1
iπ z−∞
∞ 1
p
2 2
(1 − cos p) e − c p t . e − ipx dp
pi
2.14 FINITE FOURIER TRANSFORMS

as
The finite Fourier sine transform of F(x), 0 < x < l is defined as
fs ( p) =
l
F( x) . sin
p πx
z
dx ; p ∈ I
g
0 l
.c
Similarly, the finite Fourier cosine transform of F(x), 0 < x < l is defined as
fc ( p) =
l
F( x) . cos
p πx
z
dx ; p ∈ I
w
0 l
Generally, the choice of π as the upper limit of integration in these transforms is found
w
convenient and can easily be arranged by having suitable substitutions to actual problems,
then
z z
w
π π
fs ( p) = F( x) . sin px dx and fc ( p) = F( x) . cos px dx.
0 0
Note. fs(p) is always zero when p = 0.
2.15 INVERSE FINITE FOURIER TRANSFORMS
Inversion formulae are given as follows:

(1) When upper limit is π
For sine transform:
∞
2
F( x) =
π ∑
p=1
fs ( p) sin px
www.cgaspirants.com
www.cgaspirants.com
For cosine transform:
F( x) =
1
π
fc (0) +
2
π
∞
∑ f ( p) cos px where f (0) stands for

p=1
c c z
0
π
F( x) dx .
(2) When upper limit is l

For sine transform:
∞
2 p πx
F( x) =
l ∑ f ( p) sin
p =1
s
l
For cosine transform:
F( x) =
1
fc (0) +
2
∞
∑ f ( p) cos
pπ x
where fc(0) stands for z l
F( x) dx.
m
c
l l p =1
l 0
o
EXAMPLES
.c
x
Example 1. Find finite Fourier sine transform of F(x) = 1 − .
Sol. The finite Fourier sine transform of F(x) is given by,
ts π
z FGH
n
π IJ x
fs ( p) =
K 1− sin px dx
ra
0 π
R x F cos px IJ UV − FG − 1IJ . FG − cos px IJ dx
= SFG 1 − IJ G −
z
π
π
TH π K H p K W H π K H p K
pi
0
0
1 F sin px I
π
1
= − G J = 1p .
as
p πp H p K 0
cos k( π − x)
Example 2. Find finite Fourier cosine transform of F(x) = − .
g
k sin kπ
z
.c
− cos k (π − x)
π
Sol. fc ( p) = cos px dx
0 k sin kπ
z
w
1 π
= – [cos { k (π – x) + px} + cos { k(π – x) – px}] dx
2k sin kπ 0
w
= –
1 LM
sin (kπ – kx + px) sin (kπ – kx – px)
–
OP π
N Q
w
2k sin kπ p– k p+k 0
=
1 1
−
1
= 2
1 LM
, k ≠ 0, 1, 2, ...
OP
2k p − k p + k p − k2 N Q
Example 3. Find F.F.S.T. and F.F.C.T. of F(x) = 2x, 0 < x < 4.
Sol. (i) fs ( p) = zL
0
4
2 x . sin
pπx
4
dx (Here l = 4)
MM R|| − cos p4πx U||OPP R| − cos pπx U| R 32

4
= 2x . S
MM | FG pπ IJ V|PP − z0
4
2.S
|
F pπ I
4 | dx |
V
|| GH 4 JK || |T
= S pπ
(1 − cos pπ) , p ≠ 0
N |T H 4 K |WQ 0 T W 0, p=0
www.cgaspirants.com
www.cgaspirants.com
(ii) fc ( p) = z
0
4
2 x . cos
pπx
4
dx
LM R sin pπx UOP R| sin pπx U|

4
|| 4 ||P
= M2 x . S
MM | FG pπ IJ V|PP
− 2.S
| 4 |V dx
z
|| FGH p4π IJK ||
0
4
N |T H 4 K |WQ T
0 W
R| − cos pπx U| 4
32 8 | 4 | = 32 (cos pπ − 1), p ≠ 0
= sin pπ − S
pπ | F pπ I |
V pπ 2 2
pπ
G J
|T H 4 K |W
m
0
When p = 0, zR
fc ( p) = fc (0) =
4
2 x dx = 16 .
o
0
|S sin2ppπ/2 , U|
.c
p = 1, 2 , ...
Example 4. Find F(x) if fc (p) =
|T π/4, V| where 0 < x < 2π.
p=0 W
Sol. F( x) =
1 π
. +
n 2
2π 4 2π
∑
p=1
∞
sin pπ/2
2p
. costs
pπx
2π
ra
∞
1 π 2 sin pπ / 2 px
= . +
2π 4 2π ∑
p =1
2p
cos
2
pi
∞
1 1 sin ( pπ/2) px
F( x) = +
8 π ∑ 2p
cos
2
.
as
p =1
ASSIGNMENT
g
.c
1. Find finite Fourier sine and cosine transforms of

(i) F(x) = x2, 0 < x < π (ii) F(x) = 1, 0 < x < π (iii) F(x) = x, 0 < x < π.
w
2. Find finite Fourier cosine transform of F( x) = 1 −

FG x IJ 2
H K .
w
π
3. Find finite Fourier sine transform of
w
x
(i) F(x) = (ii) F(x) = sin nx, n ∈ I (iii) F(x) = x(π2 – x2)
π
(iv) F(x) = x (π – x) (v) F(x) = ecx (vi) cos mx.
4. Find finite Fourier cosine transform of
(i) F( x) =
π
−x+
x2
(ii) sin nx, n ∈ I (iii) F( x) =
1, 0 < x < π / 2
.
UV
3 2π − 1, π / 2 < x < π W
R|x, 0 ≤ x ≤ π / 2 U|
5. Find finite Fourier sine transform of F( x) = S|π − x, π ≤ x < πV| .
T 2 W
6. Find inverse finite Fourier sine transform of
2π (− 1) p− 1 1 − cos pπ
(i) fs ( p) = , p = 1 , 2 , ... (0 < x < π) (ii) fs ( p) = , 0 < x < π.
p3 p2 π 2
www.cgaspirants.com
www.cgaspirants.com
Answers
R|− π (− 1) + 2 [(− 1)
2 p
U R| 2π(− 1) , p ≠ 0U|
p
p ≠ 0| | |V
p
− 1], 2
1. (i) (a) f ( p) = S p p 3
V (b) f ( p) = S p
s
|T 0, p = 0|W
c
|| , p = 0||
π 3
T 3 W
R| 1 {1 − (− 1) p}, p ≠ 0U| R0, if p ≠ 0UV
(ii) (a) f ( p) = S p V (b) f ( p) = S
s
|T 0, p = 0|W
c
Tπ, if p = 0W
R| {(− 1) p − 1} , p ≠ 0U|
R| π (− 1) p+1, p ≠ 0U| | p2 |V
(iii) (a) f ( p) = S p V (b) f ( p) = S
s
|T 0, p = 0|W
c
|| π , 2 |
m
p = 0|
T 2 W
R 2 , p ≠ 0U
o
2.
| 2
f ( p) = S πp
|V
|| , p = 0||
.c
c
π
T3 W
3.
(− 1) p+ 1
(i)
2
p
(iv) 3 [1 − (− 1) ]
p
(v)
0, if p ≠ n
(ii) π / 2, if p = n
p
n
[1 − (− 1) p eπc ]
(iii)
(vi)
UV
W ts 6π
p3
(− 1) p+ 1
p
[1 − (− 1) p cos mπ]
p p + c2
2
p − m2
2
ra
1 2n
4. (i) 2 (ii) 2 ; (n − p) is odd and 0 if even
p n − p2
pi
2 pπ 2 pπ
(iii) sin ; p ≠ 0 and 0 if p = 0 5. 2 sin
p 2 p 2
F 1 − cos pπ I
as
∞ ∞
2 2π(− 1) p− 1 2
6. (i)
π ∑
1
p3
sin px (ii)
π3 ∑ GH p JK sin px.
1
2
g
2.16 PARSEVAL’S IDENTITY FOR FOURIER TRANSFORMS

.c
M.A. Parseval (1755–1836), a French Mathematician, gave the following result.

w
If the Fourier transforms of F(x) and G(x) are f(p) and g(p) respectively, then
1 ∞
z z
w
(i) f ( p) g ( p) dp = ∞ F( x) G ( x) dx
2π – ∞ –∞
z z
w
1 ∞ ∞
(ii) | f ( p)|2 dp = |F( x)|2 dx
2π − ∞ −∞
where bar implies the complex conjugate.
Proof. z ∞
F( x) G ( x) dx = z ∞
F( x)
RS 1 z ∞ UV
g ( p) e ipx dp dx
Using inversion formula
−∞ −∞ T 2π −∞ W for Fourier transform
=
1
2π z−∞
∞
g ( p) RSz
T−∞
∞
F( x) e ipx dxUV dp
W
|Changing order of integration
=
1
2π z−∞
∞
f ( p) g ( p) dp ...(1) | By definition
www.cgaspirants.com
www.cgaspirants.com
Now, take G(x) = F(x) in (1), we get

1
2π z −∞
∞
f ( p) f ( p) dp = z ∞
−∞
F( x) F( x) dx
⇒
1 ∞
2π – ∞
Hence the results.

z 2
| f ( p)| dp =
∞
– ∞
| F( x)|2 dx z ...(2)
Corollary 1. Following Parseval’s identities for Fourier cosine and sine transforms can be
proved as above:
(i)
2
z ∞
f c ( p) gc(p) dp =
z ∞
F( x) G(x) dx (ii)
2
z ∞
|f c ( p)|2 dp = z ∞
|F( x)|2 dx
m
π 0 0 π 0 0
(iii)
2
z ∞
f s ( p) gs(p) dp = z ∞
F( x) G ( x) dx (iv)
2
z ∞
|f s ( p)|2 dp = z ∞
| F( x)|2 dx .
o
π 0 0 π 0 0
.c
EXAMPLES
Example 1. Using Parseval’s identity, show that

n ts z
0
∞ x2
(a + x ) (b 2 + x 2 )
2 2
dx =
π
2 (a + b)
ra
Hence find z0
∞ x2
( x 2 + 1) 2
dx . (U.P.T.U. 2015)
pi
p
Sol. If F(x) = e– ax then fs(p) =
as
a + p2 2
∴ By Parseval’s identity for sine transform,
z z
g
2 ∞ ∞
f s ( p) g s ( p) dp = F( x) G ( x) dx
.c
π 0 0
⇒
2
z ∞ p
⋅
p
dp =
z ∞
e – ax . e – bx dx
w
π 0 a 2 + p 2 b2 + p 2 0
LM OP
z
∞
w
∞ p2 π e
– ( a + b) x
π
⇒ dp = =
0 (a 2 + p2 )(b2 + p2 ) MN
2 – (a + b) PQ 2(a + b)
w
0
Thus,
z ∞
(a 2 + x 2 ) (b2 + x 2 )
0
x2
dx =
π
2(a + b)
For a = b = 1, we get
z ∞
0 ( x + 1)2
x2
2
dx =
π
4
z
2
∞ sin ax π (1 − e − a )
Example 2. Using Parseval’s identity, show that dx = .
0 x( a 2 + x 2 ) 2 a2
www.cgaspirants.com
www.cgaspirants.com
Sol. Let F(x) = e–ax and G(x) =

RS1, 0< x<a UV
T0, x>a W
a sin ap
then fc (p) = 2 2
and gc(p) =
a +p p
Parseval’s identity for Fourier cosine transform is
2
π z ∞
0
f c ( p) gc (p) dp = z
0
∞
F( x) G ( x) dx
z z
2
2 ∞ a sin ap ∞
1 – e–a
⇒ dp = e – ax . 1 dx =
π 0 p ( a 2 + p2 ) 0 a
m
∞ sin ax π 2
∴ dx = 2
(1 – e – a )
2 2
0 x( a + x ) 2a
o
.c
ASSIGNMENT
1. Using Parseval’s identity, show that z ∞

0
dx
( a2 + x 2 ) (b2 + x 2 ) ts =
π
2ab (a + b)
.
z
n
∞ dx
Hence evaluate
0 ( x 2 + 1) 2
ra
2. If F(x) =
RS1,
T0,
| x| < a
| x| > a
UV
, prove that
W z ∞ sin 2 ax
x2
dx =
πa
2
using Parseval’s identity.
pi
3. Evaluate using Parseval’s identity:
zFG 1 – cos x IJ dx
z LMHint: Take F(x) = 1, OP
as
2
∞ ∞ sin 4 x 0≤ x≤1
(i)
0 H x K (ii)
0 x 2
dx
N 0, x≥1 Q
z
FG sin x IJ 4
LMHint: Take F( x) = RS1 –| x|, UVOP
g
∞ | x|< 1
dx = π/3
4. Prove that:
H x K
0
N T 0, otherwise WQ
.c
5. Use Parseval’s identity to prove that: z ∞ ( x cos x – sin x) 2

dx =
π
w
6 15
0 x
Answers
w
w
1. π/4 3. (i) π/2 (ii) π/2
2.17 THE Z-TRANSFORM
The Z-transform plays an important role in the field of Communication Engineering and Control
Engineering at the stage of analysis and representation of discrete-time linear shift invariance
system. When continuous signals are sampled, discrete-time functions arise. The application
of Z-transform in discrete time systems is similar to that of the Laplace transform in continuous
time systems.
www.cgaspirants.com
www.cgaspirants.com
2.18 DEFINITIONS
2.18.1. One-Sided Z-transform

Let {f(k)} be a sequence defined for all positive integers ‘k’. Then the Z-transform of f(k) is
defined as
∞
Z{f(k)} = F(z) =
k=0
∑ f (k) z −k
...(1)
where z is an arbitrary complex number and Z is an operator of Z-transform.

This is one-sided Z-transform.
2.18.2. Two-Sided Z-transform
m
If {f(k)} is a sequence defined for k = 0, ± 1, ± 2, ....
∞
o
then Z{f(k)} = F(z) = ∑ f ( k) z − k ...(2)
.c
k=−∞
where z is an arbitrary complex number and Z is an operator of Z-transform.
Note
This is two sided Z-transform.
1. If f(k) = 0 for k < 0 then {f(k)} is called a casual sequence.
n ts −1
2. If f(k) is a non-casual sequence, f(k) = 0 for k ≥ 0, then its Z-transform is F(z) = ∑ f (k) z −k
and
ra
−∞
is also called one sided Z-transform.
3. The curly bracket { } represents sequence. Sequence {f(k)} is an ordered list of real or complex
pi
numbers.
4. The infinite series on R.H.S. of (1) will be convergent only for certain values of z depending on
as
sequence {f(x)}.
5. The inverse Z-transform of Z{f(k)} = F(z) is defined as Z–1[F(z)] = {f(k)}.
g
2.19 UNIT STEP AND UNIT IMPULSE SEQUENCES

.c
Unit step sequence is defined as 1

w
u(k) =
RS1, k≥0 UV
T0, k<0 W
w
Unit impulse sequence is defined as

w
–2 –1 0 1 2 3 k
δ(k) =
RS1, k=0 UV 1
T0, k≠0 W δ(k)
–2 –1 0 1 2 3 k
2.20 RELATION BETWEEN UNIT IMPULSE SEQUENCE AND UNIT STEP

SEQUENCE
∞
u(k) = ∑ δ(k)
k=−∞
and δ(k) = u(k) – u(k – 1)
www.cgaspirants.com
www.cgaspirants.com
We have δ(n – k) =
RS1, k=n UV
T0, k≠n W
∞
Also, f(n) = ∑
k=−∞
f (k) δ (n − k) .
2.21 Z-TRANSFORM OF UNIT IMPULSE FUNCTION
We know that δ(k) =

1, k = 0 UV
0, k ≠ 0 W
m
∞
∴ Z{δ(k)} = ∑ δ(k) z −k
= [...... + 0 + 0 + 1 + 0 + 0 + ......] = 1
o
k=−∞
Hence, Z{δ(k)} = 1
.c
2.22 Z-TRANSFORM OF DISCRETE UNIT STEP FUNCTION
We know that u(k) =

n
1, k ≥ 0 UV ts
0, k < 0 W
ra
∞ ∞
∴ Z{u(k)} = ∑ u( k) z − k = ∑z −k
pi
k=−∞ k=0
1 z
as
= 1 + z–1 + z–2 + ...... = −1 = ; | z | > 1.

1− z z−1
2.23 Z-TRANSFORM FOR DISCRETE VALUES OF t

g
.c
If f(t) is a function defined for discrete values of t, where t = nT, n = 0, 1, 2, ......, T being the
sampling period, then Z-transform of f(t) is defined as
w
∞
Z{f(t)} = ∑ f (nT)z −k
= F( z ).
w
k=0
Note. The important element of discrete-time systems is the samples in which a switch close to admit
w
an input signal in every T seconds. A samples is a conversion device which converts a continuous signal
into a sequence of pulses occurring at sampling instants 0, T, 2T, ...... where T is the sampling period.
EXAMPLES
Example 1. Find the Z-transform of the following sequences:
(i) f(k) = {15, 10, 7, 4, 1, − 1, 0, 3, 6 } (ii) f(k) = {5, 6, 1, 2, – 1, 0, 8, 4, 3}
A
(iii) f(k) =
RS 1 UV (iv) f(k) =
RS 1 UV , – 3 ≤ k ≤ 3
T4 W
k
T2 Wk
(v) f(k) =
RS5 ,
k
k<0
.
UV
T2 ,
k
k≥0 W
www.cgaspirants.com
www.cgaspirants.com
Sol. (i) The symbol ↑ is used to denote the term in zeroth position i.e., when k = 0. k is
an index of position of a term in a sequence.
1 1 3 6
Z{f(k)} = 15z3 + 10z2 + 7z + 4 + − 2 + 0 + 4 + 5
z z z z
1 1 3 6
⇒ F(z) = 15z3 + 10z2 + 7z + 4 + − 2 + 4 + 5
z z z z
(ii) In case the symbol ↑ is not given, extreme left term is considered as zeroth term
corresponding to k = 0. Here, the zeroth term is 5.
6 1 2 1 8 4 3
∴ Z{f(k)} = 5 + + 2 + 3 − 4 + 0 + 6 + 7 + 8
z z z z z z z
6 1 2 1 8 4 3
⇒ F(z) = 5 + + 2 + 3 − 4 + 6 + 7 + 8
m
z z z z z z z
∞ ∞
1
(iii) Z{f(k)} = ∑
f (k) z − k = z−k ∑
o
k
k=−∞ k=−∞
4
.c
1 1 1
= ...... + 64z3 + 16z2 + 4z + 1 + + 2
+ + ......
4 z 16 z 64 z 3
(iv) Z{f(k)} =
3
−3
3
∑ f (k) z
n −k
ts (since – 3 ≤ k ≤ 3)
1 1 1 1
ra
= ∑2
−3
k
z − k = 8z3 + 4z2 + 2z + 1 + +
2z 4 z 2
+ 3
8z
−1
pi
∞
(v) Z{f(k)} =
k=−∞
∑ 5k z−k + ∑
k=0
2k z− k
LM OP
as
2 4 8
= [...... + 5–3z3 + 5–2z2 + 5–1z] + 1 + + + + ......
N z z2 z3 Q
g
5 −1 z 1 z z
= + = + ; | z | < 5, | z | > 2.
.c
−1
1 − 5 z 1 − (2 / z) 5 − z z − 2
w
Example 2. Find the Z-transform of

1 1
w
(i) f(k) = {a| k |} (ii) f(k) = ,k≥1 (iii) f(k) = ,k≥0

k k(k + 1)
kπ 0, k > 0 UV
w
(iv) f(k) = cos ,k≥0 (v) f(k) = .

2 1, k ≤ 0 W
∞
Sol. (i) Z{f(k)} = ∑
k=−∞
f (k) z − k
−1 ∞ F aI
∞
+ az) + ∑ GH z JK
k
=
k=−∞
∑ a −k
z −k
+ ∑
k=0
k
a z −k
= (...... + a3 z 3 + a 2z 2
k=0
az a FG IJ −1
az z ∵ (1 − x) −1 =
∞
∑ xk
=
1 − az
+ 1−
z H K = +
1 − az z − a
.
k=0
www.cgaspirants.com
www.cgaspirants.com
F 1I
ZG J=∑
∞
1 −k
(ii)
H kK k=1
k
z
1 1 1 1 1 FG IJ
= + 2 + 3 + ...... = – log 1 −
z 2z 3z z
if
z
<1 H K
zFG IJ
= log
z−1 H K if | z | > 1.
RS 1 UV = Z FG 1 − 1 IJ = Z FG 1IJ − Z FG 1 IJ
T k(k + 1) W H k k + 1K H kK H k + 1K
(iii) Z
F z IJ − ∑ 1 z = log F z I − FG 1 + 1 + 1
= log G
∞
IJ
GH z − 1JK H 2z 3z
−k
m
H z − 1K k+1 k=0
2
+ ......
K
F z IJ − zR|S 1 + 1 FG 1IJ + 1 FG 1IJ + ......U|V 2 3
o
= log G
H z − 1K |T z 2 H z K 3 H z K W|
.c
F z I R F I 1 U
H z − 1JK − z ST− log GH 1 − z JK VW
= log G
F z IJ − z log FG z IJ = (1 – z) log FG z IJ .
= log G
H z − 1K
n
H z − 1K H z − 1K
ts
ra
F kπ IJ = ∑ cos kπ z
Z G cos
∞
−k
(iv)
H 2K k=0
2
pi
1 1 FG 1 IJ −1
z2
= 1−
z2
+
z4
− ...... ∞ = 1 +
H z2 K = 2
z +1
if | z | > 1.
as
∞ ∞
1
(v) Z{f(k)} = ∑ f (k) z − k = ∑ zk = ; if | z | < 1.
g
k=−∞ k=0
1− z
.c

w
R| ak U| 1
(i) f(k) = S k ! , k≥0
V (ii) f(k) = , k ≥ 0.
|T0, otherwise|W
w
(k + 1) !
w
∞ ∞
ak −k (az −1 ) k
∑ ∑
−1
Sol. (i) Z{f(k)} = z = = e az = ea/z.
k=0
k! k=0
k!
(ii)
L 1 OP = ∑ 1 z = 1 + 1 z + 1 z + ......
ZM
∞
–k –1 –2
N (k + 1) ! Q (k + 1) !
k=0 2! 3!
L 1 z + 1 z + ......OP
= z Mz + −1 −2 −3
N 2! 3! Q
L
= z M1 + z +
1
z +
1 O
z + ...... − 1P
−1 −2 −3
N 2 ! 3 ! Q
−1
= z (ez – 1) = z(e1/z –1).
www.cgaspirants.com
www.cgaspirants.com

(i) u(k – 1) (ii) 4 k δ(k – 1) ; k ≥ 0
(iii) δ(k – n) ; k ≥ 0 (iv) { nCk} ; 0 ≤ k ≤ n. (M.T.U. 2014)
∑
∞
1 . z−k =
1 1 1
+ 2 + 3 + ...... =
1 1 1 FG IJ
Sol. (i) Z{f(k)} =
k=1
z z z z
1 + + 2 + ......
z z H K
1 1 1
= if <1
z FG 1 − 1IJ z
H zK
1
= if | z | > 1.
m
z−1
∞
4
(ii) Z{f(k)} = ∑ 4k δ(k – 1) z–k = .
o
k=0
z
.c
∞
1
(iii) Z{f(k)} = ∑ δ(k – n) z–k =
zn
, n is (+)ve integer.
(iv) Z{f(k)} =
k=0
∑
n
k=0
nC
k
n ts
z–k = 1 + nC1 z–1 + nC2 z–2 + ...... + nCn z–n = (1 + z–1)n.
ra
Example 5. Determine the Z-transform of the sequence given by

R| U|
pi
|| 21, k
k<0
||
f(k) = S G J
F I k
k = 0, 2, 4, ...... V
as
|| H 2 K
,
||
|T FGH 13 IJK
k
, k = 1, 3, 5,...... |
g
W
.c
What is the region of convergence for the Z-transform F(z)?

∞
w
Sol. F(z) = ∑
k=−∞
f (k) z–k
w
∑ FGH 2 IJK ∑ FGH 3 IJK

−1 ∞ k ∞ k
1 1
= ∑ 2k z–k + z–k + z–k
w
k=−∞ k=0 k=0

(k-even) (k-odd)
∑2
∞
−m m
∞
F 1I
+ ∑G J
2p
−2 p F 1I
∞
+ ∑ G J
2q + 1
z − (2 q + 1)
=
m=1
z
H 2K
p=0
z
H 3K
q=0
where m = − k,
k k−1
p= ,q=
2 2
= F1(z) + F2(z) + F3(z)
z/2 z2 z/3
= + +
1 − z / 2 z2 − 1 1
z2 −
4 9
www.cgaspirants.com
www.cgaspirants.com
Region of convergence for F1(z): | z | < 2

1
Region of convergence for F2(z): | z | >
2
1
Region of convergence for F3(z): | z | >
3
1
Hence, Region of convergence for F(z): < | z | < 2.
2
ASSIGNMENT
m
1. Determine the Z-transform of the following sequences :
(i) f(k) = {2, 4, 5, 7, 0, 1, 2} (ii) f(k) = {3, 1, 2, 5, 7, 0, 1}
A A
o
(iii) f(k) = {0, 0, 1, 2, 5, 4, 0, 1} (iv) f(k) = {1, 2, 5, 4, 0, 1}
.c
(v) f(k) = δ(k + n).
2. Find the Z-transform of
f(k) = 2,
0,
1,
R|
S|
n for k < 0
for 0 ≤ k ≤ 5
for 6 ≤ k ≤ 10 .
U|
V| ts
3, T for k > 10 W
ra
3. Find the Z-transform of

R| F 1I k
G J
pi
(iii) S H 3 K
k≥0
(i) u(k – 4) (ii) δ(k – 5) .
|T(− 2) k
k≤−1
as
4. Determine Z-transform for the sequences given below :

f(0) = 1 ; f(1) = 4.7 ; f(2) = 0 ; f(3) = 0
g
f(4) = 0.75 ; f(5) = 2 ; f(k) = 0 ; k ≥ 6.

.c
5. Express the signals shown below in terms of unit impulse functions and hence find the Z-trans-
form.
w
(i) f(k) (ii) f(k)

w
2
2 1 2
1
1
w
0.5
–1 3
k k
–2 –1 0 1 2 –2 0 1 2
–1 – 0.5
Answers
7 1 2 7 1
1. (i) 2z2 + 4z + 5 + + 3 + 4 (ii) 3z3 + z2 + 2z + 5 + + 3
z z z z z
(iii) z–2 + 2z–3 + 5z–4 + 4z–5 + z–7 (iv) 1 + 2z–1 + 5z–2 + 4z–3 + z–5 (v) zn
2. 1 + z + z + z + z + z + 2(z + z + z + ...... + z ) + 3(z + z–12 + ......)
–1 –2 –3 –4 –5 –6 –7 –8 –10 –11
www.cgaspirants.com
www.cgaspirants.com
z −4 z
−
z
3. (i) ;|z|>1 (ii) z–5 (iii) 1
1− z −1
z− z+2
3
4. F(z) = 1 + 4.7 z–1 + 0.75 z–4 + 2 z–5
5. (i) f(k) = 0.5 δ(k + 1) + 2δ(k) + 2δ(k – 1) + δ(k – 2); F(z) = 0.5z + 2 + 2z–1 + z–2
(ii) f(k) = δ(k + 2) – δ(k + 1) + 2δ(k) + δ(k – 2) – 0.5 δ(k – 3); F(z) = z2 – z + 2 + z–2 – 0.5 z–3.
2.24 PROPERTIES OF Z-TRANSFORMS
2.24.1. Linearity Property
m
Z{a f(k) ± bg(k)} = a Z{f(k)} ± b Z {g(k)}
o
Proof. Z{a f(k) ± b g(k)} = ∑ { a f (k) ± b g (k)} z − k | by definition
.c
k=−∞
∞ ∞
∑ ∑
=a
k=−∞
f (k) z − k ± b
k=−∞
ts
g (k) z − k = a Z{f(k)} ± b Z{g(k)}.
Remark. If Z{f(k)} = F(z) and Z{g(k)} = G(z), then Z–1 [a F(z) ± b G(z)] = a Z–1 {F(z)} ± b Z–1 {G(z)}
n
where a and b are constants and Z–1 is inverse Z-transform operator.
ra
2.24.2. Change of Scale Property
FG z IJ
pi
Z{|ak f(k)|} = F
If Z{f(k)} = F(z), then
H aK
as
∞
Proof. F(z) = Z{f(k)} =
k=−∞
∑ f (k) z −k
g
z
.c
Substituting for z, we get

a
zFG IJ ∞
FG z IJ −k ∞
w
F
a H K ∑
=
k=−∞
f (k)
H aK = ∑a
k=−∞
k
f (k) z − k = Z{|ak f(k)|}.
w
z/a z
Remark (i) Z{ak U(k)} = = if | z | > | a |
z
−1 z−a
w
a
z z
(ii) Z{ak} = since Z(1) = .
z−a z−1
2.24.3. Multiplication by kn
If Z{f(k)} = F(z), then Z{kn f(k)} = − z FG d IJ n

F(z)
H dz K
∞
Proof. Z{k f(k)} = ∑ k f (k) z
k=−∞
−k
www.cgaspirants.com
www.cgaspirants.com
∞ ∞
d −k d
=−z
k=−∞
∑ f (k) (− kz −k−1
) =−z ∑ f ( k)
dz
(z ) = − z
dz
F( z)
k=−∞
In general, Z{|kn f(k)|} = − z

FG d IJ n
F( z ) .
H dz K
Remark. Z{(k ak)} = − z
d
[Z(a k )] = − z
d z
=
az FG .
IJ
dz dz z − a H
( z − a) 2 K
z
Note. Since Z(1) = , we have the following results.
z−1
z z( z + 1)
m
(i) Z(k) = (ii) Z(k2) =
( z − 1)2 ( z − 1)3
z3 + 4 z 2 + z z4 + 11z3 + 11z2 + z
o
(iii) Z(k3) = 4 (iv) Z(k4) = .
( z − 1) ( z − 1)5
.c
2.24.4. Shifting Property
If Z{f(k)} = F(z), then Z{f(k ± n)} = z±n F(z)

n
∞
ts ∞
Proof. Z{f(k ± n)} = ∑ f (k ± n) z − k = z ± n ∑ f (k ± n) z − ( k ± n)

ra
k=−∞ k=−∞
∞
= z±n ∑ f (r) z −r
pi
(r = k ± n)
k=−∞
= z±n F(z).
as
Corollary 1. For casual sequence,

Z{f(k – n)} = z–n F(z)
g
Also, Z{f(k + 1)} = z F(z) – z f(0)

.c
and Z{f(k + 2)} = z2 F(z) – z2f(0) – z f(1)

RS
f (1)
= z 2 F( z) – f (0) –
and so on.
UV
w
T
z W
Corollary 2. Z–1[z–n F(z)] = f(k – n) = Z–1[F(z)]k → k – n.
w
2.24.5. Division by k
w
If Z{f(k)} = F(z), then Z

RS f(k) UV = −
TkW z z
z −1 F(z) dz
Proof.
R f (k) UV = ∑
ZS
∞
f ( k) − k
z
TkW k=−∞
k
= ∑
∞
k=−∞
f ( k)
FG 1 z IJ = − ∑
Hk K
−k
∞
k=−∞
f (k) z z
z − k − 1 dz
=− z z ∞
∑ f (k) z
k=−∞
−k−1
dz
www.cgaspirants.com
www.cgaspirants.com
F I
=− z z
z –1
H
∞
GG ∑ f (k) z
k=−∞
−k
JJ dz = –
K z z
z −1 F( z) dz .
2.24.6. Initial Value Theorem
If Z{f(k)} = F(z), k ≥ 0, then f(0) = Lt F(z)

z→∞
∞
Proof. Z{f(k)} = ∑ f (k) z
k=0
−k
⇒ F(z) = f(0) + f(1) z–1 + f(2)z–2 + ......
m
Taking limit as z → ∞, we get
f(0) = Lt F( z) .
o
z→∞
.c
2.24.7. Final Value Theorem
If Z{f(k)} = F(z) ; k ≥ 0, then Lt f(k) = Lt (z − 1) F(z)
∞
n k→∞
ts z→1
Proof. Z{f(k + 1) – f(k)} = ∑ { f (k + 1) − f (k)} z − k

ra
k=0
n
z F(z) – f(0) – F(z) = Lt ∑ { f (k + 1) − f (k)} z − k
pi
n→∞
k=0
n
∑ { f (k + 1) − f (k)} z −k
as
Lt ( z − 1)F( z) = f (0) + Lt . Lt
z→1 z→1 n→∞
k=0
n
∑
g
= f (0) + Lt Lt { f (k + 1) − f (k)} z − k
n→∞ z→1
.c
k=0
| Changing the order of limits
LM O
w
n
= Lt f (0) + ∑ { f (k + 1) − f (k)}P
n→∞ MN PQ
w
k=0
= Lim f (n + 1) = Lim f (n) = Lim f ( k) .
n→∞ n→∞ k→∞
w
2.24.8. Differentiation Property

Let Z[{f(k)}] = F(z). An infinite series can be differentiated term by term within its region of
convergence. F(z) may be treated as a function of z–1.
∞ ∞
F(z) = ∑
k=0
f(k) z–k = ∑ f(k) (z–1)k
k=0
Differentiating on both sides w.r.t. z–1
∞
d
dz −1
F(z) =
k=0
∑
kf(k) (z–1)k–1 ...(1)
z–1
dF( z)
dz −1
= ∑
k=0
kf(k) z–k = Z{kf(k)}
www.cgaspirants.com
www.cgaspirants.com
dF( z)
∴ Z{kf(k)} = z–1 ...(2)
dz −1
Differentiating (1) w.r.t. z–1 again, we get
∞
d 2 F( z)
d( z −1 ) 2
= ∑
k=0
k(k – 1) f(k) (z–1)k – 2
2 ∞
d F( z)
z–2
d( z −1 ) 2
= ∑
k=0
k(k – 1) f(k) z–k = Z{k(k – 1) f(k)}
d 2 F( z)
m
∴ Z[k(k – 1) f(k)] = z–2 ...(3)
d( z −1 ) 2
o
2.24.9. Convolution Theorem (U.P.T.U. 2015)
.c
The convolution of two sequences {f(n)} and {g(n)} is defined as
∞
w(n) =
k=−∞
∑
n f(k) g(n – k) = f ∗ g
ts
If it is one sided (right) sequence, take f(k) = 0, g(k) = 0 for k < 0, then
ra
∞
w(n) = ∑ f (k ) g(n – k) = f * g.
k=0
pi
Statement. If w(n) is the convolution of two sequences f(n) and g(n), then
Z{w(n)} = W(z) = Z{f(n)} . Z{g(n)} = F(z) G(z) (U.P.T.U. 2015)

as
LM ∞ OP LM ∞ ∞ OP
MN ∑ f ( k) g ( n − k)
PQ ∑ MN ∑ f ( k) g ( n − k)
Proof. W(z) = Z{w(n)} = Z = –n
PQ z
g
k=0 n=0 k=0

.c
∞ L ∞ O
= ∑ f(k) M ∑ g (n − k) z P
−n
MN PQ
w
k=0 n=0
(by changing the order of summation)
w
∞ L ∞ OP
= ∑ f(k) M ∑ g ( p) z
− ( p+ k)
MN PQ , (putting n – k = p)
w
k=0 p=0
L ∞ OL ∞ O
= M ∑ f (k) z P M ∑ g ( p) z P = F(z) G(z).
−k −p
MN k=0
PQ MN p=0
PQ
Note. This result will be true only for the values of z inside the region of convergence.
2.24.10. Another form of Convolution Theorem

If Z{f(t)} = F(z), Z{g(t)} = G(z), then the convolution product is
∞
w(t) = ∑
k=0
f(kT) g(nT – kT) = f * g
and Z{w(t)} = W(z) = Z{f(t)} Z{g(t)} = F(z) G(z).
www.cgaspirants.com
www.cgaspirants.com
Proof. (Here we are dealing with one sided Z-transform only)

∞ ∞
F(z) = ∑
m=0
f(mT) z–m ; G(z) = ∑ g(nT) z–n
n=0
R| ∞ U| R| g(nT) z
∞ U|
F(z) G(z) = S ∑ f (mT) z
|T
−m
V| S| ∑ V|−n
m=0 WT n=0 W
∞ ∞ F
∞ n I
= ∑ ∑ f(mT) g(nT) z–m–n = ∑ GG ∑ f ( pT) g{(n − p)T}JJ z –n
n=0 m=0 H
n=0 p=0 K
m
R| U
S| ∑ f ( pT) g{(n − p)T}|V| z
∞
=Z –n = Z{f ∗ g}.
T W
o
p=0
.c
2.24.11. Time Reversal Property
If F(z) is the Z-transform of f(k), then Z{f (– k)} = F(z–1)
2.24.12. Correlation Property

If Z{f1(k)} = F1(z) and
n
Z{f2(k)} = F2(z), then Rf
ts
(z) = F1(z). F2(z–1)
1 f2
ra
and cross-correlation sequence rf1 f2 (l) = Z–1[ Rf1 f2 (z)].

pi
2.25 SOME IMPORTANT Z-TRANSFORM RESULTS

as
S. No. {f(k)}, k ≥ 0 F(z) = Z {f(k)}

g
1. δ(k) 1
.c
z
2. U(k) or 1
z−1
w
z
3. k ,|z|>1
( z − 1)2
w
FG − z d IJ n
4. kn f(k)
H dz K F(z)
w
kn
FG − z d IJ n
z
5.
H dz K z−1
,|z|>1
z
6. ak or ak U(k) ;|z|>|a|
z−a
1 FG z IJ , | z | > 1
7.
k
log
H z − 1K
8.
1
z log G
F z IJ
k+1 H z − 1K
www.cgaspirants.com
www.cgaspirants.com
1
9. e1/z
k!
10. δ(k – n) z–n
11. f(0) Lt F(z)

z→∞
12. Lt f(k) Lt (z – 1) F(z)

k→∞ z→1
∞
13. rf1 f2 (l) =
k=−∞
∑ f1(k) f2(k – l) R f1 f2 (z) = F1(z) F2(z–1)
m
14. h(k) = f(k) ∗ g(k) F(z) . G(z)
15. f(– k) F(z–1)
o
.c
EXAMPLES
Example 1. Determine the Z-transform of

n
f(k) = δ(k + 1) + 3 δ(k) + 6δ(k – 3) – δ(k – 4).
ts
ra
Sol. By linearity property, we have
F(z) = Z{f(k)} = Z{δ(k + 1)} + 3 Z{δ(k)} + 6 Z{δ(k – 3)} – Z{δ(k – 4)}
= z + 3 + 6z–3 – z–4.
pi

RS (k + 1) (k + 2) UV
as
(i) {ak+3} (ii) {k2} (iii)

T 2 W
(iv) {abk} ; a ≠ 0, b ≠ 0 (v) {k (k – 1)}.
g
∞
.c
Sol. (i) Z{f(k)} =

k=−∞
∑a k+3
z − k = a3 Z(ak) …(1)
w
z
We know that, Z(1) =
z−1
w
z/a z
∴ Z(ak) = = (By change of scale property)
w
( z/a) − 1 z − a
za3
∴ From (1), Z{f(k)} = .
z−a
(ii) FG
Z(k2) = − z
d IJ 2
Z(1) = − zFG d IJ LMFG − z d IJ FG z IJ OP
H dz K H dz K NH dz K H z − 1K Q
=− z
d LM
z
=
z( z + 1) .OP
N
dz ( z − 1) 2
( z − 1) 3 Q
(iii) Z
RS (k + 1)(k + 2) UV = 1 Z(k 2
+ 3k + 2)
T 2 W 2
www.cgaspirants.com
www.cgaspirants.com
1 1 z( z + 1)
+3 −z
d
Z(1) +
LM
2z RS UV OP
[Z(k2) + 3 Z(k) + Z(2)] =
=
2 2 ( z − 1) 3
dz N
z−1 T W Q
=
LM
1 z( z + 1)
− 3z
d z
+
2z
=
FG
1 z( z + 1) IJ
+
3z
+
2z OP LM OP
N
2 ( z − 1) 3
dz z − 1 z−1 H
2 ( z − 1) 3
K( z – 1) 2
z−1
.
Q N Q
(iv) Z(abk) = a Z(bk) = a
FG z IJ
H z − bK if | z | > | b |.
z ( z + 1) z 2z
(v) Z{k(k – 1)} = Z(k2 – k) = Z(k2) – Z(k) = 3
− 2
= .
( z − 1) ( z − 1) ( z − 1) 3
Example 3. Find the Z-transform of f(k) = u(– k).
m
z
Sol. We know that Z{u(k)} =
z−1
o
By time reversal property, we have
.c
z −1 1
Z{u(– k)} = −1 = ; | z | < 1.
z −1 1− z
(i) {k + nCn} or {k + nCk} ; k ≥ 0
n ts
(ii) {k + nCn ak} ; k ≥ 0.
Sol. (i) Z{k+nCn} = Z{k+nCk}
ra
∞
= ∑ k+nC
k z–k = 1 + n+1C1 z–1 + n+2C
2 z–2 + ......
pi
k=0
(n + 2)(n + 1) –2
= 1 + (n + 1) z–1 + z + ......
as
2!
(− n − 1)(− n − 2)
= 1 + (– n – 1)(– z–1) + (– z–1)2 + ......
2!
g
F z IJ n+1
.c
=G
= (1 – z–1)–n–1 = (1 – z–1)–(n+1)
H z − 1K .
w
F z / a IJ
=G
n+ 1
(ii) Z{k+nC ak }
H z / a − 1K | Change of scale property using (i)
w
F z IJ .
=G
n +1
w
H z − aK
(i) sin αk, k ≥ 0 (U.P.T.U. 2009, 2015) (ii) sin (3k + 5), k ≥ 0.
∞ ∞
1
Sol. (i) Z(sin αk) = ∑
k=0
sin αk z–k =
2i ∑
k=0
(eiαk – e–iαk) z–k
1 LM ∞ ∞ OP
MN ∑ (e ∑ (e
iα
= z –1 ) k − − iα
z –1 ) k
2i
k=0 k=0
PQ
1
= [(1 – eiα z–1)– 1 – (1 – e–iα z–1)–1]
2i
www.cgaspirants.com
www.cgaspirants.com
=
1 LM 1 −
1 OP = 1 LM z −
z OP
2i N1 − e z iα −1
1− e − iα
z −1
Q 2i N z − e iα
z−e − iα
Q
=
1 LM z(e iα
− e − iα ) OP = z sin α
.
2i N z − z(e
2 iα
+e − iα
) + 1Q z 2
− 2 z cos α + 1
∞ ∞
1
(ii) Z{sin (3k + 5)} = ∑ sin (3k + 5) z–k =
2i
∑
k=0
{ei(3k + 5) – e–i(3k + 5)} z–k
k=0
1 LM ∞ ∞ OP
=
2i MN ∑
k=0
e5i (e 3i z –1 ) k − ∑
k=0
e −5i (e −3i z –1 ) k
PQ
m
1 5i
= [e (1 – e3i z–1)–1 – e–5i (1 – e–3i z–1)–1]
2i
o
e 5iF 1 I F
e −5i 1 I
= GH JK – GH JK
.c
2i 1 − e z 3 i −1
2i 1 − e −3 i z −1
e F z I
5i
F z −5i
I=1 LM e 5i
z( z − e −3i ) − e −5i z( z − e3i ) OP
= G J
2i H z − e K
–
e
2i GH z − e
n
1 L (e − e ) z − z(e − e
5i
3i
−5i 2 2i
−3i
ts
− 2i
JK 2i
)O 2
N 3i
(z − e ) (z − e −3i
) Q
= M 1 − z (e + e ) + z PQ = zz sin 5 − z sin 2
; | z | > 1.
ra
2i N 3i −3i 2 2
− 2 z cos 3 + 1
pi
(i) ck cosh (α k), k ≥ 0 (ii) ck cos (α k), k ≥ 0

FG kπ + αIJ , k ≥ 0.
(iii) cosh
H2 K (U.P.T.U. 2008)
as
∞
Fe αk
+ e − αk Iz 1 LM ∞ ∞ OP
∑ GH JK MN ∑ (e z −1 ) k + ∑ (e z −1 ) k
g
–k α −α
Sol. (i) Z{cosh (αk)} = =
2 2 PQ
.c
k=0 k=0 k=0
=
1
(1 − e α z −1 ) −1 + (1 − e − α z −1 ) −1 =
1 z
+
z LM OP
w
2 2 z−e α
z − e− α N Q
LM OP
w
z 2 z − (eα + e − α ) z ( z − cosh α)
. 2= = 2
α
2 z − z(e + e ) + 1 N
−α
z − 2 z cosh α + 1 Q
w
By change of scale property,

z z FG IJ
Z{ck cosh (αk)} =
c c H
− cosh α
K =
z ( z − c cosh α)
.
FG z IJ − 2 FG z IJ cosh α + 1
2
z − 2cz cosh α + c 2
2
H cK H cK
∞
F e + e I z = 1 LM (e
i αk − iαk ∞ ∞ OP
2 M∑ ∑ (e
Z{cos (αk)} = ∑ G
H 2 JK
iα
(ii) –k z −1 ) k + − iα
z −1 ) k
k=0 N k=0 k=0
PQ
=
1
(1 − e iα z −1 ) −1 + (1 − e − iα z −1 ) −1 =
1 z
+
z LM OP
2 2 z−e iα
z − e − iα N Q
www.cgaspirants.com
www.cgaspirants.com
zF 2 z − 2 cos α z ( z − cos α) I
= GH2
2 z − 2 z cos α + 1
= 2
z − 2 z cos α + 1
JK
By change of scale property,
FG z IJ FG z − cos αIJ
k
Z{c cos (αk)} =
H cK H c K = z (z − c cos α)
FG z IJ − 2 FG z IJ cos α + 1 z − 2cz cos α + c
2 2 2
H cK H cK
R F I U ∞
F I R| FGH ∞ kπ IJ
K
FG kπ + αIJ U
H 2 K |V z–k
∑ e
1 +α −
kπ
T H 2 JK VW = ∑ cosh GH 2 + αJK z = 2 S|T
kπ
(iii) Z Scosh G
2
+α –k +e
|W
m
k=0 k=0
1 L O
∞ ∞
o
= M
2 MN
e ∑ (e
α π/2
z ) + e ∑ (e
–1 k
z ) P
−α
PQ
−π /2 –1 k
.c
0 0
1
= [eα (1 – eπ/2 z–1)–1 + e–α (1 – e–π/2 z–1)–1]
2
LM F
1 α z
n
F z I OP = z LM e (z − e ) + e (z − e ) OP
I ts α −π/2 −α π/2
MN GH JK
GH z − e JK PQ 2 N (z − e ) (z − e ) Q
= e . + e–α
2 z−e π/2 −π /2 π/2 −π /2
ra
L F π IO
z M z cosh α − cosh G − αJ P
L O N H 2 KQ .
pi
z 2 z cosh α − cosh (α − π/2)

= M
2 N z − (2 cosh π/2) z + 1 Q
2 P =
z − 2 z cosh π/2 + 1 2
R F kπ + αIJ UV ; k ≥ 0.
as
Example 7. Find the Z-transform of Scos G

T H 8 KW
g
R F kπ + αIJ UV = ∑ cos FG kπ + αIJ z = ∑ FG cos kπ cos α − sin kπ sin αIJ z

Sol. Z Scos G
∞ ∞
T H 8 KW H8 K H 8
–k
K
–k
.c
k=0 8k=0
F kπ IJ – sin α Z FG sin kπ IJ
w
= cos α Z G cos H 8K
H 8K
w
R| U F I
z ( z − cos π / 8) | G z sin π / 8 JJ
= cos α . S V – sin α . G
w
|T z − 2z cos 8 + 1|W
π 2
GH z − 2 z cos 8 + 1JK
π 2
| From Ex. 5 and 6

FG z
π 2 π IJ
z 2 cos α − z cos
π FG IJ
=
H8
− z cos
cos α − z sin sin α
8
=
K 8
−α
.
H K
π π
z 2 − 2 z cos + 1 z 2 − 2 z cos + 1
8 8
z(z − cos aT)
Example 8. If F(z) = 2 , find f(0).
z − 2z cos aT + 1
Sol. From initial value theorem, f(0) = Lt F(z)
z→∞
www.cgaspirants.com
www.cgaspirants.com
z( z − cos aT)
∴ f(0) = Lt 2 = 1 by L’ Hospital rule.
z→∞ z − 2 z cos aT + 1
z
Example 9. If F(z) = , find Lt f(t).
z − e−T t→∞
Sol. From final value theorem, we have
Lt f(t) = Lt (z – 1)F(z) = Lt (z – 1) z
= 0.
t→∞ z→1 z→1 z − e −T
Example 10. Find Z–1

RS 1 UV given Z RS z UV = (– 1) . –1 k
T z + 1W T z + 1W
m
Sol. Z S
R 1 UV = Z RSz . z UV
–1 –1 −1
T z + 1W T z + 1W
o
=Z S
R z UV = R|SFG − 1IJ U|V k
.c
|TH K |W
–1 = {(– 1) }, k = 1, 2, 3, ...... k–1
T z + 1W k→ k − 1 k→ k− 1
Example 11. Find Z G

F 3 IJ .
H 3z − 1K
–1
n ts
LM R UOP
ra
R 3 U R| 1 U| | z |VP
Sol. Z S –1
V =Z –1 = Z Mz S –1 −1
T 3z − 1W S| z − V| 1 M | z − |P
1
NM T 3 WQP
pi
T 3W
R z U| = FG 1IJ or F 1I u(k – 1).
as
k− 1 k− 1
=Z | GH 3 JK
–1
S| z − 1 V| H 3K
T 3W
g
k→ k − 1
.c
Example 12. Using differentiation property, find the Z-transform of

(i) k ak u(k) (ii) k(k – 1) ak u (k).
w
d FG z IJ
Sol. (i) Z{kak u(k)} = z–1
dz −1 H z − aK
w
d a az −1
= z–1 (1 – az–1)–1 = z–1 . =
w
−1
dz (1 − az ) −1 2
(1 − az −1 ) 2
d2 2 a 2 z −2
(ii) Z {k(k – 1) ak u(k)} = z–2 (1 – az–1)–1 = .
d( z −1 ) 2 (1 − az −1 ) 3
z −1 2 z −2
Note. If a = 1, then Z{k u(k)} = −1 2 and Z{k(k – 1) u(k)} = .
(1 − z ) (1 − z −1)3
Example 13. Find the Z-transform of f ∗ g where
(i) f(n) = u(n), g(n) = 2n u(n)
(ii) f(n) = 3n u(n), g(n) = 4n u(n) using convolution theorem.
∞
z
Sol. (i) F(z) = Z{u(n)} = ∑
n=0
1 . z–n =
z−1
if | z | > 1
www.cgaspirants.com
www.cgaspirants.com
∞
z
G(z) = Z{2n u(n)} = ∑
n=0
2n z–n =
z−2
if | z | > | 2 |
By convolution theorem
Z{f ∗ g} = Z{w(n)} = W(z) = F(z). G(z)
z z z2
= . = if | z | > | 2 |.
z−1 z−2 ( z − 1)( z − 2)
z
(ii) F(z) = Z{3nu(n)} = if | z | > | 3 |
z−3
z
G(z) = Z{4nu(n)} = if | z | > | 4 |
m
z−4
By convolution theorem
Z{f ∗ g} = Z{w(n)} = W(z) = F(z) . G(z)
o
z z z2
.c
= . = if | z | > | 4 |.
z−3 z−4 ( z − 3)( z − 4)
Example 14. Compute the convolution f(k) of the two sequences :
n
f1(k) = {4, – 2, 1} and
ts
f2(k) =
RS1,
T0,
0≤ k≤ 5
otherwise
UV
.
W
ra
Sol. F1(z) = Z{f1(k)} = 4 – 2z–1 + z–2

F2(z) = Z{f2(k)} = 1 + z–1 + z–2 + z–3 + z–4 + z–5
pi
∴ F(z) = F1(z) . F2(z) = 4 + 2z–1 + 3z–2 + 3z–3 + 3z–4 + 3z–5 – z–6 + z–7
Taking inverse Z-transform, we get
as
f(k) = {4, 2, 3, 3, 3, 3, − 1, 1} .
A
g
Example 15. Determine the cross-correlation sequence rf1 f 2 (l) of the sequences :
.c
f1(k) = {1, 2, 3, 4 } and f2(k) = {4, 3, 2, 1}.

Sol. Cross-correlation sequence can be obtained using the correlation property of
w
Z-transform.
F1(z) = 1 + 2z–1 + 3z–2 + 4z–3
w
F2(z) = 4 + 3z–1 + 2z–2 + z–3

⇒ F2(z–1) = 4 + 3z + 2z2 + z3
w
R f1 f2 (z) = F1(z) . F2(z–1) = (1 + 2z–1 + 3z–2 + 4z–3) (4 + 3z + 2z2 + z3)

= (z3 + 4z2 + 10z + 20 + 25z–1 + 24z–2 + 16z–3)
∴ rf1 f2 (l) = Z–1{ R f f (z)} = {1, 4, 10, 20, 25, 24, 16} .
1 2
A
2
2z + 5z + 14
Example 16. If F(z) = , evaluate f(2) and f(3).
(z − 1) 4
Sol. F(z) =
1
.
LM 2 + 5z + 14 z OP
−1 −2
z 2
N (1 − z ) Q
−1 4
By initial value theorem, f(0) = Lt . F(z) = 0

z→∞
www.cgaspirants.com
www.cgaspirants.com
Similarly, f(1) = Lt z{F(z) – f(0)} = 0

z→∞
f(2) = Lt z2{F(z) – f(0) – z–1f(1)} = 2 – 0 – 0 = 2

z→∞
f(3) = Lt z3{F(z) – f(0) – z–1f(1) – z–2 f(2)} = Lt z3{F(z) – 2z–2}

z→∞ z→∞
LM 2z + 5z + 14 − 2 OP
2
= Lt z3 .
z→∞ N (z − 1) z Q
4 2
R 13z + 2z + 8 z − 2 UV
3 2
= Lt z .S
3
z→∞ T z (z − 1) W = 13.
2 4
m
ASSIGNMENT
o
.c
1. Find the Z-transform (one sided) of the following sequences {f(k)} where f(k) is
(i) FG 1IJ k
u(k) (ii) (cos θ + i sin θ)k (iii) (– 1)k u(k)
H 4K
kπ
(iv) 3k sin
2
(v) 2k cos
kπ
2
n ts (vi) 4k +
FG 1IJ
H 2K
k
+ u(k – 3).
ra
2. Find the Z-transform of {f(k)} where f(k) = k 2k
3. Show that
z( z − cosh θ) z sinh θ
pi
(i) Z(cosh kθ) = (ii) Z(sinh k θ) = .

z2 − 2 z cosh θ + 1 z2 − 2 z cosh θ + 1
4. Show that
as
ze a ( ze a − cos θ) ze a sin θ
(i) Z(e–ak cos kθ) = (ii) Z(e –ak sin kθ) = .
z 2 e 2 a − 2 ze a cos θ + 1 z 2 e 2 a − 2 ze a cos θ + 1
g
z( z + 1) z2 ( z + 1)
Using Z(n2) = , show that Z(n + 1)2 =
.c
5. 3 .
( z − 1) ( z − 1)3
2 z2 + 3 z + 4
w
6. Given Z(un) = , | z | > 3, show that u1 = 2, u2 = 21, u3 = 139.

( z − 3)3
w
z ( z3 + z) cos θ − 2 z2
7. Using Z(k) = , show that Z(k cos kθ) = .
( z − 1)2 ( z2 − 2 z cos θ + 1)2
w
8. Find the convolutions of

kπ kπ
(i) k(k – 1) ∗ 3k (ii) 3k ∗ cos kθ (iii) cos ∗ sin .
2 2
d
9. Prove that Z{kn} = – z [Z{kn–1}].
dz
∞ ∞
10. Evaluate the Z-transform of the sequence {f(k)} = ∑
k=0
2k ∑
k=0
3k.
Answers
4z z z 3z
1. (i) (ii) (iii) (iv)
4z − 1 z−e iθ z+1 2
z +9
www.cgaspirants.com
www.cgaspirants.com
z2 z 2z 1
(v) 2 (vi) + + 2
z +4 z − 4 2 z − 1 z ( z − 1)
2z 2 z2 z2 ( z − cos θ)
2. 8. (i) (ii)
( z − 2)2 ( z − 1)3 ( z − 3) ( z − 3)( z2 − 2 z cos θ + 1)
z3 z2
(iii) 10. .
( z + 1)2
2
( z − 2)( z − 3)
2.26 INVERSE Z-TRANSFORM
Inverse Z-transform is a process for determining the sequence which generates given
Z-transform. If F(z) is the Z-transform of the sequence {f(k)}, then {f(k)} is called the inverse
Z-transform of F(z). The operator for inverse Z-transform is Z–1.
m
If Z{f(k)} = F(z), then Z–1 [F(z)] = {f(k)}.
o
2.27 METHODS OF FINDING INVERSE Z-TRANSFORMS
.c
We have the following methods of finding inverse Z-transforms:
(1) Convolution method (2) Long division method
(3) Partial fractional method
(5) Power series method.
ts
(4) Residue method (or Inverse Integral method)
n
ra
2.27.1. Convolution Method
We know that Z{f ∗ g} = F(z) G(z)
k
pi
∴ Z–1 {F(z) G(z)} = f ∗ g = ∑ f (m) g(k − m) .

m=0
as
EXAMPLES
g
RS z UV using convolution theorem.

2
.c
Example 1. Find Z–1

T (z − a)(z − b) W
w
–1
Sol. We know that Z {F(z) G(z)} = f ∗ g
z R z UV = a
∴ f(k) = Z {F(z)} = Z S
w
Let F(z) = –1 –1 k
(z − a) Tz − a W
R z UV = b
w
∴ g(k) = Z {G(z)} = Z S
z –1 –1 k
G(z) =
z−b Tz − bW
Z–1{F(z) G(z)} = f ∗ g = ak ∗ bk
∑ FGH ab IJK
k k m
= ∑
m=0
ambk–m = bk
m=0
(a G.P.)
R| FG a IJ − 1U|
k+1
= bk
|S H b K |V = a k+ 1
− b k+ 1
.
|| b − 1 ||
a a−b
T W
www.cgaspirants.com
www.cgaspirants.com
R| z2 U|
Example 2. Using Convolution theorem, evaluate Z–1 (z − 1)(z − 3) . S| V|
T W
Sol. We know that
Z–1 {F(z) . G(z)} = f ∗ g
z
Let F(z) = ∴ f(k) = (1)k
z−1
z
G(z) = ∴ g(k) = (3)k
z−3
Now, Z–1 {F(z) . G(z)} = (1)k ∗ (3)k
∑ FGH 31IJK
k k m
= ∑ 1m 3k–m = 3k (a G.P.)
m
m=0 m=0
R| FG 1IJ − 1U|
k+ 1
|S H 3 K
o
= 3k
|V = (1) k+ 1
− (3) k+ 1
=
RS
1 k+ 1
(3 − 1) .
UV
|| 3 − 1 || T W
.c
1 1− 3 2
T W
2.27.2. Long Division Method
n
∞
ts
Since Z-transform is defined by the series F(z) = ∑ f(k) z–k (one sided), to find the inverse
ra
k =0
Z-transform i.e., Z–1 [F(z)], expand F(z) in the proper power series and collect the coefficient of
pi
z–k to get f(k).
EXAMPLES
as
Example 1. Find inverse Z-transform of

g
10z 2(z 3 − z)
.c
(i) 2 (ii) .
z − 3z + 2 (z 2 + 1) 2
10 z −1
w
10 z
Sol. (i) F(z) = 2 =
z − 3z + 2 1 − 3 z −1 + 2 z −2
w
By actual division,
10z–1 + 30z–2 + 70z–3 + ......
w
1 – 3z–1 + 2z–2 ) 10z–1

10z–1 – 30z–2 + 20z–3
30z–2 – 20z–3
30z–2 – 90z–3 + 60z–4
70z–3 – 60z–4
70z–3 – 210z–4 + 140z–5
+ 150z–4 – 140z–5
∴ F(z) = 10z–1 + 30z–2 + 70z–3 + ......
www.cgaspirants.com
www.cgaspirants.com
Now comparing the quotient with

∞
∑
k=0
f(k) z–k = f(0) + f(1)z–1 + f(2) z–2 + ......
We get the sequence f(k) as f(0) = 0, f(1) = 10, f(2) = 30, f(3) = 70, ......
i.e., we can get f(k) = 10(2k – 1), k = 0, 1, 2, ......
2 z( z 2 − 1) 2 z −1 − 2 z − 3
(ii) F(z) = =
( z 2 + 1) 2 1 + 2 z − 2 + z −4
By actual division, we get F(z) = 2z–1 – 6z–3 + 10z–5 – 14z–7 + ......
m
∞
Comparing the quotient with ∑
k=0
f(k) z–k = f(0) + f(1)z–1 + f(2)z–2 + ......
o
We get f(0) = 0, f(1) = 2, f(2) = 0, f(3) = – 6, f(4) = 0, f(5) = 10, f(6) = 0, ......
.c
kπ
In general f(k) = 2k sin
, k = 0, 1, 2, ......
2
Example 2. Find the inverse Z-transform of
F(z) =
4z
z−a
n ts
ra
for (i) | z | > | a | (ii) | z | < | a |.

Sol. (i) For | z | > | a |, we have
pi
4z 4z FG 1 − a IJ −1
∑ FGH az IJK
∞ k
a
z−a
=
z H zK =4 if
z
<1
as
k=0
∞
= ∑ 4ak z–k, where | z | > | a |
g
k=0
.c
Z–1
FG 4z IJ = {4a }. k
H z − aK
w
(ii) For | z | < | a |, we have

w
4z 4z FG 1 − z IJ −1
z−a
=–
a H aK
w
4z F z z 2
z3 I
=–
a H G 1+ +
a a 2
+
a 3 JK
+ ...... , if |z|<|a|
4z 4 z2 4 z3 4 z4
=– – 2 − 3 − 4 – .....
a a a a
Z–1
FG 4z IJ = {f(k)}
H z − aK
where
RS
{f(k)} = ......−
4
,
−4 −4 −4
, , .
UV
T a 4
a3 a2 a W
www.cgaspirants.com
www.cgaspirants.com
Example 3. Find the inverse Z-transform of (U.P.T.U. 2014)

1
F(z) = for
(z − 3)(z − 2)
(i) | z | < 2 (ii) 2 < | z | < 3 (iii) | z | > 3.
1 1 1
Sol. F(z) = = −
( z − 3)( z − 2) z−3 z−2
(i) For | z | < 2.
1 z FG 1 IJ
z
−1
FG IJ −1
F(z) = –
3
1−
3 H +
2
1−
2K H K
1 1
= – (1 + 3–1z + 3–2z2 + 3–3z3 + ...) + (1 + 2–1z + 2–2z2 + 2–3z3 + ...)
m
3 2
= – (3–1 + 3–2z + 3–3z2 + ...) + (2–1 + 2–2z + 2–3z2 + ...)
o
Here coeff. of z–k (if k > 0) = 0
1 1
.c
Coeff. of z–k (if k < 0) = – − k + 1 + − k+ 1 = – 3k–1 + 2k–1
3 2
Now, Z–1{F(z)} = {f(k)} = {– 3k–1} + {2k–1}.
(ii) For 2 < | z | < 3.
n
FG IJ – 1 FG 1 − 2 IJ
1 z
−1
ts −1
F(z) = –
H K z H zK
1−
ra
3 3
1 F z z z I 1F 2 22 3 2
23 I
=– G
3 H
1+ + + + ......J – G 1 + +
K zH z z + JK
+ ......
pi
2 3 2
3 3 3 z3
= – (3–1 + 3–2z + 3–3z2 + 3–4z3 + ......) – (z–1 + 2z–2 + 22z–3 + ......)
as
Here, coeff. of z–k (if k > 0) = – 2k–1

Coeff. of z–k (if k ≤ 0) = – 3k–1
g
R|− 2 k−1
, k>0 U|
S|− 3 V
.c
Now, Z–1{F(z)} = {f(k)} =

T k−1
, k ≤ 0 |W
w
(iii) For | z | > 3.

1 FG 3 1IJ −1
2 FG IJ −1
w
F(z) =
z
1−
H z
–
z K1−
z H K
w
1 1
= (1 + 3z–1 + 32z–2 + ......) – (1 + 2z–1 + 22z–2 + ......)
z z
∴ Z–1{F(z)} = {f(k)} =
3 k− 1 RS
−2 k− 1
, k≥1
.
0,T k≤0
2.27.3. Partial Fractional Method

Here we split the given F(z) into partial fractions whose inverse transforms can be
written directly.
Example. Find the inverse Z-transform of
z z 3 − 20z 8z 2
(i) 2 (ii) (iii) .
z + 7z + 10 (z − 2) 3 (z − 4) (2z − 1)(4z − 1)
www.cgaspirants.com
www.cgaspirants.com
F( z) 1 1 A B
Sol. (i) = 2 = = +
z z + 7 z + 10 ( z + 2)( z + 5) z+2 z+5
1 1
− .
1 1 1 1 FG IJ
=
3 z+2 3 z+5
∵ A = ,B= −
3 3 H K
1 z 1 z
∴ F(z) = −
3 z+2 3 z+5
∴ f(k) = Z–1{F(z)} =
1 –1
Z
RS z UV − 1 Z RS z UV −1
3 T z + 2W 3 T z + 5W
=
RS 1 (– 2) k –
1
(– 5)k
UV LM∵ Z(a k ) =
z OP
T3 3 W N z−a Q
m
z 3 − 20 z F( z) z 2 − 20
(ii) F(z) = or =
( z − 2) 3 ( z − 4) ( z − 2) 3 ( z − 4)
o
z
.c
F( z) z 2 − 20 A + Bz + Cz 2 D
Now = 3 = 3
+
z ( z − 2) ( z − 4) ( z − 2) z−4
⇒ D=–
1
2
, A = 6, B = 0, C =
n
1 2
1
2
1 FG IJ
ts
−
H K
ra
6+ z
F( z) 2 2
∴ = 3
+
z ( z − 2) z − 4
pi
or F(z) =
RS 12z + z − z UV = 1 RS z(z − 2) + 4 z + 8z − z UV
1 3 2 2
T (z − 2) z − 4 W 2 T (z − 2)
2 3
z − 4W 3
as
1 R z 2z + 4 z z U
2
= S
2 Tz − 2
+ 2.
( z − 2)
− V
z − 4W 3
g
LM∵ Z R az + a z U = k a OP
.c
2 2
1
Now –1
f(k) = Z {F(z)} = {2 + 2k 2 – 4 },
2
k 2 k
MN ST ( z − a) VW
k
PQ
−1
3
2 k
w
= {2k–1 + 2k . k2 – 22k–1}
w
2
8 z2 z
(iii) F(z) = =
(2 z − 1)(4 z − 1) 1 FG
1 IJ FG IJ
w
z−
2
z−
4H KH K
F( z) z A B 2 1
= = + = −
z FG z − 1IJ FG z − 1IJ z−
1
z−
1
z−
1
z−
1
H 2K H 4K 2 4 2 4
2z z
∴ F(z) = −
z − (1 / 2) z − (1 / 4 )
R| FG 1IJ − FG 1IJ U| , k = 0, 1, 2, ......
k k
f(k) = Z–1{F(z)} = 2 S| H 2 K H 4 K V|
T W
www.cgaspirants.com
www.cgaspirants.com
2.27.4. Inverse Integral Method (or Residue Method)

By using the theory of complex variables, it can be shown that the inverse Z-transform is
given by f(k) =
1
2πi C z
F(z) zk–1 dz, where C is the circle (may be even closed contour) which
contains all the isolated singularities of F(z) and containing the origin of the z-plane in the
region of convergence. Hence by Cauchy’s Residue theorem,
f(k) = sum of the residues of the singularities of F(z).
EXAMPLES
m
Example 1. By Residue method, find the inverse Z-transform of
z z z2 + z
o
(i) (ii) (iii)
z 2 + 7z + 10 z 2 − 2z + 2 (z − 1)(z 2 + 1)
.c
2z z(z + 1) z(z 2 − 1)
(iv) (v) (vi) .
z3 − z 2 + z − 1
Sol. (i) F(z) = 2

z
z + 7 z + 10
n (z − 1) 3
ts (z 2 + 1) 2
z
ra
1
f(k) = zk–1 F(z) dz = sum of residues
2πi C
z z
pi
1 z 1 zk
= zk–1 2
dz = dz
2πi
z + 7 z + 10
C 2πi C ( z + 2)( z + 5)
as
Poles are z = – 2, – 5. These are simple poles.

zk (− 2) k
Residue (at z = – 2) = Lt (z + 2) =
g
z→ −2 ( z + 2)( z + 5) 3
.c
zk (− 5) k
Residue (at z = – 5) = Lt (z + 5) =
( z + 2)( z + 5) −3
w
z→ −5
∴ f(k) = sum of residues =

(− 2) k
+
RS
(− 5) k UV = 1 {(– 2) k – (– 5)k}.
|T |W 3
w
3 −3
z
w
(ii) F(z) = 2
z − 2z + 2
f(k) =
1
2πi C z
zk–1 F(z) dz = sum of the residues
=
1
2πi C
zk–1 . 2
z
z
z − 2z + 2
dz =
1 zk
2πi C z 2 − 2 z + 2
dz z
The poles are given by z2 – 2z + 2 = 0 ∴ z = 1 ± i both are simple poles.
Residue at z = 1 + i is
zk (1 + i) k
Lt [z – (1 + i)] =
z→ 1+ i [ z − (1 + i)][ z − (1 − i)] 2i
www.cgaspirants.com
www.cgaspirants.com
Similarly, residue at z = 1 – i is
zk (1 − i) k
Lt [z – (1 – i)] =
z→ 1− i [ z − (1 + i)][ z − (1 − i)] − 2i
(1 + i) k k
∴ f(k) = sum of the residues = + (1 − i)
2i − 2i
RS 1 [(1 + i) – (1 – i) ]UV
= k k ...(1)
T 2i W
(1 + i) = ( 2 ) RScos
kπ kπ U
We know that k
T 4 + i sink
V
4 W
m
(1 – i) = ( 2 ) RScos
kπ kπ U
k
T 4
k
− i sin
4 W
V
o
(1 + i) – (1 – i) = ( 2 ) RS2i sin
kπ U
∴ k k
T V
4 W
k
.c
R
Substituting in (1), we get f(k) = S( 2 ) sin
kπ U
V k
(iii)
Tn
F(z) =
4 W
z2 + z
ts
( z − 1)( z 2 + 1)
ra
f(k) =
1
2πi z C
zk–1 F(z) dz = sum of the residues
pi
=
1
z zk–1
z2 + z
2 dz =
1
z zk
( z + 1)
( z − 1)( z 2 + 1)
dz
as
2πi C ( z − 1)( z + 1) 2πi C
Poles are given by z = 1, ± i

( z + 1)
g
Residue (at z = 1) = Lt (z – 1) zk =1
( z − 1)( z 2 + 1)
.c
z→1
z k ( z + 1) i k (i + 1) i k (1 + i) 1
w
Residue (at z = i) = Lt (z – i) = = = – ik
z→i ( z − 1)( z − i)( z + i) (i − 1)(2i) −2(1 + i) 2
w
1
Similarly, residue (at z = – i) is = – (– i)k
2
w
∴ f(k) = sum of the residues

1 k 1 1
i – (– i)k = 1 – {ik + (– i)k}
=1– ...(1)
2 2 2
kπ kπ kπ kπ
ik = cos + i sin ; (– i)k = cos – i sin
2 2 2 2
kπ
ik + (– i)k = 2 cos . Substituting in (1), we get
2
f(k) = 1 –
1
2 cos
kπ RS
= 1 – cos
UV RS
kπ
.
UV
2 2 T 2W T W
www.cgaspirants.com
www.cgaspirants.com
2z 2z
(iv) F(z) = 3 2 =
z − z + z−1 ( z − 1)( z 2 + 1)
Poles are z = 1, z = ± i
2zk
Residue (at z = 1) = Lt (z – 1) =1
z→1 ( z − 1)( z 2 + 1)
2z k 2i k ik
Residue (at z = i) = Lt (z – i) = =–
z→i ( z − 1)( z − i)( z + i) (i − 1)(2i) 1+ i
(− i) k
Residue (at z = – i) = –
1− i
RS i + (− i) UV
m
k k
f(k) = sum of the residues = 1 –
|T 1 + i 1 − i |W
F i I = 1 – 2R.P. of
o
k
e ik π / 2
= 1 – 2 R.P. of G
H 1 + i JK
.c
2 e iπ / 4
2 R.P. of e ik π / 2 − iπ / 4
=1–
n
=1– 2 cos
ts
FG kπ − π IJ = RS1 – FG cos kπ + sin kπ IJ UV .
H 2 4K T H 2 2 KW
ra
z( z + 1)
(v) F(z) =
( z − 1) 3
pi
z = 1 is a pole of order 3
1 d2 RS (z − 1) . z (z + 1) UV
3 k
as
∴ Residue (at z = 1) = Lt
2 ! z → 1 dz 2 |T (z − 1) |W 3
1
g
= Lt {k(k + 1)zk–1 + k(k – 1)zk–2}

2! z→1
.c
1 2
{k + k + k2 – k} = k2
=
w
2
∴ f(k) = sum of the residues = {k2}, k = 0, 1, 2, ......
w
z( z 2 − 1)
(vi) F(z) =
( z 2 + 1) 2
w
z = ± i are poles and each is a pole of order 2.
Residue (at z = i) = Lt
d RS(z − 1) 2 z k ( z 2 − 1) UV = Lt
d RS z (z − 1) UV
k 2
z→i dz T| ( z − i) 2 ( z + i) 2 W| z→i dz T| (z + i) W| 2
( z + i) 2 [ z k . 2 z + kz k− 1 ( z 2 − 1)] − z k ( z 2 − 1) . 2( z + i) k k–1
= Lt = i
z→i ( z + i) 4 2
k
Similarly, R2 = Residue of f(z) (at z = – i) = (– i)k–1
2
∴ f(k) = sum of the residues =

RS k [i k− 1 UV
+ (− i) k− 1 ] , k = 0, 1, 2, .....
T2 W
www.cgaspirants.com
www.cgaspirants.com
Example 2. Using residue’s method, evaluate Z–1

RS 8z − z UV . 3
T (4 − z) W 3
(8 z − z 3 )
Sol. Here, F(z) =
(4 − z) 3
Poles are given by, z = 4 (pole of order 3)
Residue of F(z) (at z = 4) is
R=
1 d2
( z
LM
− 4) z .
RS
3 k − 1 8z − z
3 UVOP
(3 − 1) ! dz 2 MN T
(4 − z) 3 WPQ z=4
LM OP LM OP
m
1 d2 1 d2
= {( z 3 − 8 z) z k − 1 } = (zk + 2 − 8zk )
2 dz N
2
Q z=4 N
2 dz 2 Q
o
z=4
.c
= [(k + 2)(k + 1)zk – 8k(k – 1)zk – 2]z = 4
2
1 1
= [(k + 2)(k + 1) (4)k – k(k – 1) (4)k]
=
2
4k 2 LM 1
2
n
k + 3 k + 2 − (k 2 − k) = (k2 + 7k + 4) (4)k – 1
OP
ts
2 N 2 Q
ra
∴ Z–1
RS 8 z − z UV = R = {(k
3
2 + 7k + 4) (4)k – 1}.
T (4 − z) W
pi
Example 3. Using residue’s method, show that

as
Z–1
RS 3z + 2 UV = 53 (.2)
2
k
+
31
( − .4) k .
T (5z − 1)(5z + 2) W 75 75
g
3z2 + 2
.c
Sol. Here, F(z) =

(5 z − 1)(5 z + 2)
w
Poles are given by (5z – 1)(5z + 2) = 0

1 −2
w
⇒ z= , which are simple poles. Consider a contour | z | = 1

5 5
LMFG z − 1IJ z . 3z + 2 OP
w
1 2
k−1
NH 5 K
Residue (at z = ) is R1 = Lt
5 z→
1 (5 z − 1)(5 z + 2) Q
5
= Lt1
LM 1 . 1 z (3z + 2)OP k−1 2
z→
5
N 5 (5z + 2) Q
1 1 FG IJ FG 3 + 2IJ = 53 (.2)
k−1
=
15 5 H K H 25 K 75 k
Residue (at z = – 2/5) is

FG z + 2 IJ . z k–1 3z2 + 2
R2 = Lt
z → − 2/5 H 5K .
(5 z − 1)(5 z + 2)
www.cgaspirants.com
www.cgaspirants.com
= Lt
LM 1 . 1 z (3z + 2)OP k− 1 2
N 5 (5z − 1)
z → − 2/5
Q
1 F 1I F − 2 I FG 12 + 2IJ = (− 1) . 2
k− 1 k k− 1
62
= . G− J . G J
5 H 3K H 5 K H 25 K (5) k+ 2
.
3
62 31
(– 1)k . 2k–1 (.2)k =
= (– .4)k
75 75
53 31
Hence, Z–1 {F(z)} = {f(k)} = sum of residues = R1 + R2 = (.2) k + (− .4) k .
75 75
2.27.5. Power Series Method
m
In this method we find the inverse Z-transform by expanding F(z) in power series as illustrated
in the following example.
o
Example. Find Z–1 log
RS FG z IJ UV by power series method.
T H z + 1K W
.c
R z UV ,
Sol.
1
F(z) = log S
T z + 1W
n
R 1 / y UV = – log (1 + y)
ts
Let z = then F(z) = log S
T 1 / y + 1W
ra
y
1 2 1 3 1 1 1 (− 1) k –k
=–y+ y – y + ...... = – + – + ······ + z
pi
z 2z 2 3
2 3 3z k
R| 0, for k = 0
{F(z)} = S (− 1)
as
k
∴ f(k) = z–1 .
|T k , otherwise
g
2.28 SOME IMPORTANT INVERSE Z-TRANSFORM RESULTS

.c
w
S.No. F(z) {f(k)} {f(k)}

where | z | > | a |, k > 0 where | z | < | a |, k < 0
w
1
1. ak–1 U (k – 1) – ak–1 U(– k)
z−a
w
z
2. ak U (k) or ak – ak
z−a
z2
3. (k + 1) ak – (k + 1) ak
( z − a)2
1
4. (k – 1) ak–2 U (k – 2) – (k – 1) ak–2 U (– k + 1)
( z − a) 2
zn 1 1
5. (k + 1) ...... (k + n – 1) ak U(k) – (k + 1) ...... (k + n – 1) ak
( z − a) n (n − 1) ! (n − 1) !
www.cgaspirants.com
www.cgaspirants.com
2.29 DIFFERENCE EQUATIONS
The inherent discrete nature of some physical phenomena gives rise to work with discrete
functions. The mathematical models in which a variable can have only discrete set of values,
give a chance to study difference equations. Difference calculus also forms the basis of difference
equations which arise in the theory of probability, in the study of electrical networks, in
statistical problems and in all situations where sequential relation exists at various discrete
values of the independent variable.
2.30 DEFINITION
A difference equation is a relation between the differences of an unknown function at one or
m
more general values of the argument.
Or
o
An equation which connects various differences of an unknown function is called a
difference equation.
.c
A difference equation is a relationship of the form
LM OP
F x, y,
N
Δy Δ2 y
,
Δx Δx 2
Δy
,
n ......,
Δn y
Δx n
f ( x + h) − f ( x)
Q
=0
ts ...(1)
ra
Let y = f(x), then =
Δx h
2
Δ y f ( x + 2h) − 2 f ( x + h) + f ( x)
2 =
pi
Δx h2
as
Δn y f ( x + nh) − ...... + ......

= ... ,
Δx n hn
where h is the interval of differencing. Hence eqn. (1) can be rewritten as
g
φ [x, f(x), f(x + h), f(x + 2h), ......, f(x + nh)] = 0 ...(2)
.c
where f(x) = y is an unknown function.

e.g. Δyk + 2yk = 0 ...(3)
w
2
Δ yk + 3Δyk + yk = 0 ...(4)
w
If we put Δ ≡ E – 1, where E is an operator called shift operator such that E{f(x)} = f(x + 1),
then eqns. (3) and (4) can be rewritten as
w
(E – 1)yk + 2yk = 0
⇒ (E + 1)yk = 0 ...(5)
and 2
(E – 1) yk + 3(E – 1)yk + yk = 0
⇒ E2yk + Eyk – yk = 0 ...(6)
Eqns. (3) and (4) may also be put as
yk+1 + yk = 0; yk+2 + yk+1 – yk = 0.
2.31 ORDER OF A DIFFERENCE EQUATION
The order of a difference equation is defined as the difference between the largest and the
smallest arguments for the function involved divided by h, the interval of differencing.
www.cgaspirants.com
www.cgaspirants.com
Consider the difference equation

yk+2 + yk+1 – yk = 0
(k + 2) − k
Order = = 2.
1
Note. While finding the order of a difference equation, it must always be expressed in a form free of Δ’s.
2.32 DEGREE OF A DIFFERENCE EQUATION
The degree of a difference equation is defined to be the highest power of f(x).
2.33 SOLUTION OF A DIFFERENCE EQUATION
m
A solution of a difference equation is any function which satisfies the given equation. The
general solution of a difference equation is defined as the solution which involves as many
o
arbitrary constants as the order of the difference equation.
The particular solution is a solution obtained from the general solution by assigning
.c
particular values to periodic constants.
Consider a difference equation
Let
yh + 1 – 2yh = 0 ; h = 0, 1, 2, ......
n
h
yh = 2 , h = 0, 1, 2, .....
ts …(1)
…(2)
The function yh defined by (2) satisfies the difference equation (1) so it is called a solution
ra
of eqn. (1). Generally eqn. (1) is satisfied by

yh = c . 2h for any constant c ...(3)
pi
The function yh given by eqn. (2) is a particular solution of eqn. (1) while function yh in
eqn. (3) is the general solution of eqn. (1).
as
Remark. A difference equation may have no solution just as in case of algebraic equation
e.g. (yh+1 – yh)2 + yh2 = – 1 is satisfied for no real-valued function y.
We will now proceed to solve difference equations with the help of Z-transforms.
g
.c
2.34 APPLICATIONS OF Z-TRANSFORMS TO DIFFERENCE EQUATIONS
Z-transform is useful in solving difference equations.

w
The given difference equation can be converted to the form y = φ(z) by taking Z-trans-
w
form on it, provided the initial values of y are known. Using inversion, we can get the value of
yk which is the solution of the given difference equations.
w
2.35 PROVE THAT
FG
Z(yk+n) = zn y − y 0 −
y1 y IJ
− . .. − nn−−11 , where Z(yk) = y .
H z z K
∞ ∞
Proof. LHS = Z (yk+n) = ∑
k=0
yk+n z–k = zn ∑
k=0
yk+n z–(n+k)
Setting m = n + k, we get
∞ LM∞ n−1 OP
z(yk+n) = zn ∑
m=n
ym z–m = zn
MN ∑ y
m=0
m z−m – ∑
m=0
ym z − m
PQ
www.cgaspirants.com
www.cgaspirants.com
LM
= zn y − y0 −
y1 y2 y
− 2 − ...... − nn −− 11 .
OP
N z z z Q
Note. For n = 1, 2, 3, ......, we have
Z(yk+1) = z( y – y0)
FG y − y y1 IJ
Z(yk+2) = z2
H 0 −
z K
FG y − y y1 y2 IJ and so on ...
Z(yk+3) = z3
H 0 −
z
− 2
z K
Note. If Z(yk) = y , then Z(yk–n) = z–n y .
m
EXAMPLES
o
Example 1. Solve by Z-transform: yk+1 + yk = 1 if y0 = 0.
.c
Sol. Take Z-transform on both sides, we get
Z(yk+1) + Z(yk) = Z(1)
∴ z( y – y0) + y =
z
z−1
n ts
z
ra
⇒ y (z + 1) = (∵ y0 = 0)
z−1
z 1 LM z − z OP
pi
∴ y = =
( z – 1)( z + 1) 2 N z − 1 z + 1Q
Take inverse Z-transform,
as
∴ yk =
1 LM FG
Z −1
z
− Z −1
zIJ FG IJ OP = 1 {1 – (– 1) }.k
2 N Hz−1 z+1K H KQ 2
g
Example 2. Solve by Z-transform: yk+2 – 3yk+1 + 2yk = 0; y0 = 0, y1 = 1.

.c
Sol. Take Z-transform on both sides,

w
Z(yk+2) – 3Z(yk+1) + 2Z(yk) = Z(0)

FG
z2 y − y0 −
y1 IJ – 3z ( y – y ) + 2 y = 0
H K
w
⇒ 0
z
(z2 – 3z + 2) y = z
w
z z z z
∴ y = = = –
z2 − 3z + 2 ( z − 1)( z − 2) z−2 z−1
Take inverse Z-transform, we get
yk = Z–1
FG z − z IJ = {2 k – 1} where k = 0, 1, 2, ....
H z − 2 z − 1K
Example 3. Solve the following difference equations using Z-transform
yk+1 – 2yk = 1; k ≥ 0, y0 = 1.
Sol. Taking Z-transform on both sides, we get
Z(yk+1) – 2 Z(yk) = Z(1)
www.cgaspirants.com
www.cgaspirants.com
z
⇒ z( y – y0) – 2 y =
z−1
z z2
⇒ (z – 2) y = +z= |∵ y0 = 1
z−1 z −1
z2
⇒ y=
( z − 1) ( z − 2)
y z 2 1
⇒ = = −
z ( z − 1)( z − 2) z − 2 z − 1
2z z
or y = −
z−2 z−1
m
FG z IJ – Z FG z IJ = 2 (2 ) – 1 = {2
o
yk = 2 Z–1
H z − 2K H z − 1K
–1 k k+1 – 1}.
.c
Example 4. Solve using Z-transform : yx+2 – 2 cos α yx+1 + yx = 0, y0 = 0, y1 = 1.
FG
z2 y − y0 −
H
n
zK
ts
IJ – 2 cos α . z ( y – y ) + y = 0
y1
0
ra
z2
FG y − 1IJ – 2z cos α . y + y = 0
⇒
H zK
pi
z z
y = 2
=
z − 2 z cos α + 1 ( z − e )( z − e − iα )
iα
as
−i F z I + i .F z I
y =
2 sin α GH z − e JK 2 sin α GH z − e JK
iα − iα
g
∴ Taking inverse Z-transform, we get

.c
−i
yx = [(eiα)x – (e–iα)x]
w
2 sin α
−i
w
= [(cos α x + i sin α x) – (cos α x – i sin α x)]

2 sin α
RS sin α x UV .
w
=
T sin α W
1 1
Example 5. Solve using Z-transform : yk + y = δ(k) + δ(k – 1)
4 k–1 3
where δ(k) is unit impulse sequence.
1 1
Z(yk) + Z(yk–1) = Z{δ(k)} + Z{δ(k – 1)}
4 3
1 –1 1 –1
⇒ y + z y =1+ z
4 3
www.cgaspirants.com
www.cgaspirants.com
1 −1 1
z 1+z+
⇒ y = 3 = 3
1 1
1 + z −1 z +
4 4
1
There is only one simple pole at z = – .
4
LM 1 O
FG −1IJ = MG z + J . z . 3 PP
F 1 I z+
k− 1
H
Residue at z =
4 K
is
MMH 4 K FG z + 1IJ P
N H 4 K PQ z=−
1
4
L . FG z + 1IJ OP = FG 1 − 1IJ FG − 1IJ = 1 . FG − 1IJ k− 1 k− 1
m
= Mz
k− 1
N H 3K Q H 3 4 K H 4 K 12 H 4 K
z=−
1
4
o
R| 1 FG − 1IJ U| . k−1
y = Residue = S V
.c
T| 12 H 4 K W|
∴ k
Example 6. Solve by Z-transform y

n 1
4
F 1I
+ y = G J ; k ≥ 0, y = 0.
H 4Kk+1 ts
(U.P.T.U. 2009)
k
k
0
ra
1 LMFG 1IJ OP k
Z(yk+1) + Z(yk) = Z
4 MNH 4 K PQ
pi
1 z
⇒ z( y – y0) + y =
4 1
z−
as
4
z
y =
∴
FG z − 1IJ FG z + 1IJ
g
H 4KH 4K
.c
LM 1 1 P
O
y
=2M
MN z – 41 z + 41 PPQ
w
–
z
w
F I
G z z J
w
y =2G
GH z – 41 z + 41 JJK
∴ –
Take inverse Z-transform on both sides,

R|F 1I − F − 1I U| .
k k
yk = 2 S|GH 4 JK GH 4 JK V|
T W
1 1
k
kπ FG IJ
Example 7. Solve: yk +
25
yk–2 =
5
cos
2 H K
(k ≥ 0) by residue method.
Sol. Take Z-transform on both sides, we get
FG 1 IJ = Z LMFG 1IJ k OP
H
Z yk +
25
yk − 2
K MNH 5 K cos kπ / 2
PQ
www.cgaspirants.com
www.cgaspirants.com
z( z − c cos α)
1 –2 z2 ∵ Z(c k cos αk) =
⇒ y+ z y= z − 2cz cos α + c 2
2
25 1
z2 + Here α = π / 2, c = 1 / 5
25
z2 z4
y = =
∴
FG z 2 1 IJ FG 1 + 1 z IJ FG z −2 1 IJ 2
H +
25 K H 25 K H 2
+
25 K
i −i
There are two poles of II order at z = and z = .
5 5
LM R U|OP
F iI d |F
| iI |
Residue G at z = J = M
H 5 K MM dz S|GH z – 5 JK . z . F 1 I V|PPP
2 4
z
m
k–1
2
MN |T GH z + 25 JK |WPQ 2
o
i
z=
5
LM R U|OP
.c
| LM FG z + i IJ (k + 3) z – z 2 FG z + i IJ OP
2
k+ 2 k+ 3
= M .S
d | z | H 5K H 5K P
MM dz | F i I V|PPP = MM
k+ 3
G z+ J
MN |T H 5 K |WPQ
n
M
2
MN z=
i
5
H 5K ts
FG z + i IJ PP
PQ
4
z=
i
5
ra
=G
F k + 2 IJ . FG 1IJ . i = FG k + 2 IJ . FG 1IJ FG cos π + i sin π IJ
k
k
k k
H 4 K H 5K H 4 K H 5K H 2 2K
pi
( k + 2) F π πI
. G cos k + i sin k J
=
4 . (5 ) kH 2 2 K
as
F − i IJ = (k + 2) FG cos kπ − i sin kπ IJ
Residue G at z =
H 5K 4 . (5 ) H k
2 2K
g
kπ (k + 2)
yk = sum of residues = cos .
.c
2 . (5 k ) 2
Example 8. Solve by Z-transform the difference equation
w
yk+2 + 6yk+1 + 9yk = 2k ; (y0 = y1 = 0). (U.P.T.U. 2010, 2014)

w
Z(yk+2) + Z(6yk+1) + Z(9yk) = Z(2k)

FG IJ + 6z ( y – y ) + 9 y = z
w
y1
⇒ H
z2 y − y0 −
z K z−2 0
z
⇒ (z2 + 6z + 9) y =
z−2
z
⇒ y =
( z − 2)( z + 3) 2
Poles are given by (z – 2)(z + =0 ⇒ z = 2, – 3 3)2
There are two poles out of which one is simple and other is double.
LM k− 1 z OP LM z OP
k
2k
= ( z − 2) z .
Residue (at z = 2)
N ( z − 2)( z + 3) 2 Q z=2
=
N (z + 3) Q
2
z=2
=
25
www.cgaspirants.com
www.cgaspirants.com
1 LM d R(z + 3) . z
2−1
z UVOP
Residue (at z = – 3) =
(2 − 1) ! MN dz T
2−1 S 2
( z − 2)( z + 3)
k− 1
2
WPQ z=−3
L d R z UOP = LM (z − 2) . kz − z OP
=M
k k− 1 k
MN dz S|T (z − 2) V|WQP N (z − 2) Q
z=−3
2
z=−3
− 5k(− 3) k− 1 − (− 3) k 1 (− 3) k
= = – k(– 3)k–1 –
25 5 25
RS 2 − k (– 3)
(− 3) k k
UV .
Hence f(k) = Sum of the residues =
25 T 25 5 k–1 –
W
m
Example 9. Using Z-transform, solve the following difference equation:
yk+2 + 4yk+1 + 3yk = 3k, given y0 = 0 and y1 = 1.
o
.c
Z(yk+2) + 4Z(yk+1) + 3Z(yk) = Z(3k)
FG y1 IJ z
⇒
⇒
H
z2 y − y0 −
z K
+ 4z( y – y0) + 3 y =
n
(z2 + 4z + 3) y – z =
z−3
z
z−3
ts
ra
z z
⇒ y = +
( z − 3)( z + 1)( z + 3) ( z + 1)( z + 3)
pi
Taking inverse Z-transform,

RS z UV + Z RS z UV
as
−1
yk = Z–1
T (z − 3)(z + 1)(z + 3) W T (z + 1)(z + 3) W
= Z–1 {P(z)} + Z–1 {Q(z)}
g
z 3k
.c
Residue of P(z) at (z = 3) = Lt (z – 3). zk–1 . =

z→3 ( z − 3)( z + 1)( z + 3) 24
w
z (− 1) k
Residue of P(z) at (z = – 1) = Lt (z + 1) . zk–1 . =
z → –1 ( z − 3)( z + 1)( z + 3) (− 8)
w
z (− 3) k
Residue of P(z) at (z = – 3) = Lt (z + 3) . zk–1 . =
( z − 3)( z + 1)( z + 3)
w
z → –3 12
z (− 1) k
Residue of Q(z) at (z = – 1) = Lt (z + 1) . zk–1 . =
z→−1 ( z + 1)( z + 3) 2
z (− 3) k
Residue of Q(z) at (z = – 3) = Lt (z + 3) . zk–1 . =
z→ −3 ( z + 1)( z + 3) −2
yk = sum of residues =
RS 3 − (− 1) + (− 3) UV + RS (− 1)
k k k k
−
(− 3) k UV
∴
|T 24 8 12 |W |T 2 2 |W
R 3 + 3 (− 1) − 5 (− 3) UV .
=S
k
k k
T| 24 8 12 W|
www.cgaspirants.com
www.cgaspirants.com
Example 10. Use Z-transform to solve the difference equation:

yk+2 – 2yk+1 + yk = 3k + 5.
Sol. We have,
yk+2 – 2yk+1 + yk = 3k + 5. ...(1)
Taking Z-transform on both sides of the given equation, we get
FG y IJ
z2 y − y0 − 1 – 2z ( y – y0) + y = Z(3k + 5) = 3 .
z
+ 5.
z
H z K ( z − 1) 2
z−1 ...(2)
Let y0 = a and y1 = b then, (2) becomes

FG b IJ 3z
+
5z
H
z2 y − a −
z K
– 2z( y – a) + y =
( z − 1) 2 z − 1
m
3z 5z
(z2 – 2z + 1) y – az2 + 2az – bz = +
( z − 1) 2 z − 1
o
3z 5z
⇒ (z – 1)2 y = + + az2 + (b – 2a) z
( z − 1) 2 z − 1
.c
3z 5z az 2 z
+ y = + + (b − 2a) ...(3)
( z − 1)
4
( z − 1) 3n ( z − 1) 2
ts
( z − 1) 2
LM 3z OP + Z LM 5z OP + Z LM az 2
+ (b − 2a) z OP
ra
−1 −1
yk = Z–1 ...(4)
N (z − 1) Q N (z − 1) Q N
4 3
( z − 1) 2 Q
3z
pi
Now, let F(z) =

( z − 1) 4
Pole is z = 1 of order 4.
as
1 LM d R(z − 1)
3
3z UVOP LM
1 d3 OP
MN dz ST
4
Residue (at z = 1) = . z k− 1 . = (3 z k )
WPQ N Q
g
3
3! ( z − 1) 4 z=1
6 dz 3 z=1
.c
1 k(k − 1)(k − 2)
= k(k − 1)(k − 2) z k−3 =
2 z=1 2
w
k( k − 1)(k − 2)
∴ Z–1 [F(z)] =
w
2
5z
Let G(z) =
w
( z − 1) 3
1 LM d R(z − 1)
2
5z UVOP 1 d2 LM OP
MN dz ST
3
Residue (at z = 1) = . . zk − 1 = (5 z k )
2! 2
( z − 1) 3
WPQ z=1
2 dz 2
N Q z=1
5 5 k (k − 1)
= [k(k – 1)zk–2]z = 1 =
2 2
5
∴ Z–1[G(z)] = k(k – 1)
2
az 2 + (b − 2a) z
Let H(z) =
( z − 1) 2
www.cgaspirants.com
www.cgaspirants.com
1 d LM R { az 2 + (b − 2 a) z} UVOP
MN ST
Residue (at z = 1) = z k− 1 ( z − 1) 2 .
(2 − 1) ! dz ( z − 1) 2 WPQ z=1
FG
d k− 1
{ az 2 + (b − 2a) z}
IJ LM d (az k+ 1
+ (b − 2 a) z k )
OP
H
= dz z K z=1
=
N dz Q z=1
= [a(k + 1)zk
+ (b – 2a) kzk–1]
z=1 = a(k + 1) + (b – 2a)k
= a + bk – ak = a(1 – k) + bk
∴ Z–1[H(z)] = a(1 – k) + bk
Now, from (4),
m
k(k − 2)(k − 1) 5
yk = + k(k – 1) + bk – a(k – 1)
2 2
o
k(k − 1)
= (k + 3) + bk – a(k – 1)
.c
2
k( k − 1)(k + 3)
= + (b – a)k + a
=
RS
T
2
k(k − 1)(k + 3)
n
2
+ C 0 + C 1k . ts UV
W
where, C0 = a and C1 = b – a.
ra
Example 11. Using the Z-transform, solve the following difference equation:
6yk+2 – yk+1 – yk = 0, y(0) = 0, y(1) = 1.
pi
Sol. Taking Z-transform on both sides of given equation, we get

as
Z(6yk+2) – Z(yk+1) – Z(yk) = Z(0)

FG
6z2 y − y0 −
y1 IJ – z( y – y ) – y = 0
⇒
H K
g
0
z
FG y − 1IJ – z y – y = 0
.c
6z2
H zK
w
(6z2 – z – 1) y = 6z
6z 6z z
w
or y = = =
2
6z − z − 1 (3 z + 1)(2 z − 1) FG 1 IJ FG
1 IJ
Hz+ z−
KH K
w
3 2
LMF I F I OP F I
or
y
=
GMG 1 JJ − GG 1 JJ P or y = 6 GG z − z JJ
6
z MMGH z − 1 JK GH z + 1 JK PP
5 5
GH z − 12 z + 13 JK
N 2 3 Q
Taking inverse Z-transform on both sides, we get
L F I F I OP
6 M G z JJ − Z GG z JJ P = 6 R|SFG 1IJ − FG − 1IJ U|V . k k
y = M Z G −1 −1
5 H 2K H 3 K |W
MN GH z − 2 JK GH z + 3 JK PPQ |T
5 M
k
1 1
www.cgaspirants.com
www.cgaspirants.com
yk+3 – 3yk+ 2 + 3yk+1 – yk = u (k); y(0) = y(1) = y(2) = 0.
Sol. We have,
yk+3 – 3yk+ 2 + 3yk+1 – yk = u(k) ...(1)
Taking Z-transform on both sides of equation (1), we get
FG
z3 y − y0 −
y1 y2 IJ – 3z FG y − y 2 y1 IJ + 3z ( y – y ) – y = z
H z
− 2
z K H 0 −
z K z−1 0
z
⇒ (z3 – 3z2 + 3z – 1) y =
z−1
z
⇒ y =
( z − 1) 4
m
LM z OP
o
yk = Z–1
N ( z − 1) Q 4
.c
z
Now, F(z) =
( z − 1) 4
Pole is z = 1 (Pole of order 4)
1
n
LM d R(z − 1)
3
ts z UVOP 1 LM d3 OP
MN dz ST
4
Residue at this pole = . z k− 1 . = (z k )
WPQ
ra
3! 3
( z − 1) 4
z=1
6 N dz 3
Q z=1
1 k(k − 1)(k − 2)
= [k(k – 1)(k – 2) zk–3]z=1 =
pi
6 6
Hence, yk =
k(k − 1)(k − 2)RS
, k ≥ 3.
UV
T W
as
6
R|F 1 I U|
g
F 1I
–G J y = SG J
k
V| , k ≥ 0, y(0) = 0.
H 3K T|H 3 K
yk
.c
k–1
W
w
1 |RSFG 1IJ k
|UV
Z(yk) – Z(yk–1) = Z
|TH 3 K |W
w
3
1 –1 z
w
⇒ y – z y =
3 1
z−
3
⇒ FG 1 − 1 z IJ−1 y =
z
⇒ y =
z2
...(1)
H 3 K z−
1 FG z − 1IJ 2
3 H 3K
Taking inverse Z-transform on both sides, we get
LM OP
yk = Z–1 MM z 2
PP ...(2)
MMN FGH z − 31IJK
2
PPQ
www.cgaspirants.com
www.cgaspirants.com
z2
Here, F(z) =
FG z − 1IJ 2
H 3K
1
Poles are z = which are of order 2.
3
LM R U|OP
1
|
M d |SFG z − 1IJ . z . z |VPP 2 2
(2 − 1) ! M dz |H
k− 1
MM | 3 K
Residue at this pole =
FG z − 1IJ |P 2
N T H 3 K |WPQ z=
1
3
L d (z )OP = LM(k + 1) z OP = k + 1
m
=M
k+ 1 k
N dz Q N Q z=
1 3 z=
1
k
o
3 3
y =S
R k + 1UV .
.c
∴ From (2),
T3 Wk k
k
F 1I
16
|RF 1 I F kπ I |U
y + GH JK y = SGH JK cos GH JK V , k ≥ 0.
|T 4k–2 2 |W
n k
ts
ra
1 F 1I kπ , k ≥ 0
y = G J cos
k
Sol. y + k
16 H 4K k–2 2
...(1)
pi
Taking Z-transform on both sides of equation (1), we get

1 –2 z2
y + z y =
as
16 1
z2 +
16
FG 1 + 1 z IJ
g
−2 z2
H 16 K y =
1
.c
z2 +
16
w
z4 z4
y = =
FG z 2 1 IJ FG z 2 1 IJ FG z 1 IJ 2
H +
KH +
K H 2
w
16 16
+
16 K
i −i 1
w
There are two poles of II order at z = and z = . Consider a contour | z | =

4 4 2
LM R U|OP LM R U|OP
|
d |F iI | M |
d | z |VP
Residue FG at z = IJ = M P
2 4 k+ 3
z
H 4 K MM dz S|GH 4 JK
i
z − . z .
FG z + 1 IJ V|PP
k− 1
= MM dz S| F i I 2 2
||PP
MN |T H 16 K |WPQ MN |T GH z + 4 JK
2
z=
i
4
WPQ z=
i
4
LM FG z + i IJ . (k + 3) z − z . 2 FG z + i IJ OP
2
k+2 k+ 3
= MM
H 4K H 4K P
MMN FG z + i IJ PP 4
H 4K PQ z=
i
4
www.cgaspirants.com
www.cgaspirants.com
LM FG z + i IJ . (k + 3) z − 2z OP k+ 2 k+ 3
=M
H 4K PP
MM FG z + IJ i
3
PPQ
MN H 4K z=
i
4
i F i I − 2 FG i IJ
. (k + 3) G J
k+ 2
F k + 3 IJ i − 2 i i
i G
k+ 3
3
k
3
k
=
2 H 4K H 4K
=
H 32 K 4 64 4 k k
FG i IJ 3
i
1 3
H 2K 8
=G
F k + 2 IJ FG i IJ = k + 2 FG cos kπ + i sin kπ IJ
k
H 4 K H 4 K (4) H 2 2K
m
k+1
F iI k+2 F kπ kπ I
Residue G at z = − J is =
H 4K
G
H cos − i sin
2K
J
o
(4) k+ 1 2
R (k + 2) . 2 cos kπ UV .
.c
∴ y = Sum of residues = S
k
T (4) k+1
2 W
Example 15. Solve using Z-transform: y – 5y = S

n
k+1
T 0; k < 0W
ts
Rsin k; k ≥ 0UV ; given that y
k 0 = 0.
ra
Sol. Take Z-transform, we get
z sin 1
z( y – y0) – 5 y = ...(1)
pi
2
z − 2 z cos 1 + 1
z sin 1
⇒ (z – 5) y = |∵ y0 = 0
as
2
z − 2 z cos 1 + 1
z sin 1
∴ y =
g
( z − 5)( z 2 − 2 z cos 1 + 1)
.c

LM z sin 1 OP
w
yk = Z–1
N (z − 5) ( z − e ) (z − e ) Q
i −i
w
z sin 1
Here F(z) =
( z − 5) ( z − e i ) ( z − e − i )
w
Consider a contour | z | = 6
Residue of F(z) at (z = 5)
z sin 1 5 k sin 1
= Lt (z – 5) . zk–1 . =
z→5 ( z − 5) ( z − e i ) ( z − e − i ) 26 − 10 cos 1
Residue of F(z) at (z = ei)
z sin 1
= Lt (z – ei) zk–1 .
i
z→e ( z − 5) ( z − e i ) ( z − e − i )
(e i ) k sin 1 e ik
= i i −i =
(e − 5)(e − e ) 2i (e i − 5)
www.cgaspirants.com
www.cgaspirants.com
Residue of F(z) at (z = e–i)

z sin 1
= Lt (z + e–i) . zk–1 .
−i
z→e ( z − 5)( z − e i )( z – e) – i
(e − i ) k sin 1 e − ik
= =
(e − i − 5)(e − i − e i ) 2i (5 − e − i )
f(k) = sum of residues
5 k sin 1 e ik e − ik
= + +
26 − 10 cos 1 2i (e − 5) 2i (5 − e − i )
i
5 k sin 1 LM OP
m
1 cos k + i sin k cos k − i sin k
= + +
N
26 − 10 cos 1 2i cos 1 + i sin 1 − 5 5 − cos 1 + i sin 1 Q
o
5 k sin 1 cos k sin 1 (5 − cos 1) sin k
= − –
.c
26 − 10 cos 1 26 − 10 cos 1 26 − 10 cos 1
RS FG 5 − cos 1IJ sin kUV
T
k
f(k) = A (5) − A cos k − A
Example 16. Solve by Z-transform:

n H sin 1 K Wts
yk+1 = 7yk + 10 xk
where, A=
sin 1
26 − 10 cos 1
.
ra
xk+1 = yk + 4xk ; y0 = 3, x0 = 2.
Sol. yk+1 = 7yk + 10xk
pi
Taking Z-transforms, we get

z( y – y0) = 7 y + 10 x
as
⇒ (7 – z) y + 10 x = – 3z ...(1)
Applying Z-transform to xk+1 = yk + 4xk, we get
g
z( x – x0) = 4 x + y
.c
⇒ y – (z – 4) x = – 2z ...(2)
w
Eliminating y from (1) and (2), we get

(z2 – 11z + 18) x = 2z2 – 11z
w
2 z 2 − 11z z z
w
x = = +
( z − 2)( z − 9) z − 9 z − 2
∴ xk = {9k + 2k} (Take inverse Z-transform)
The equation is xk+1 = yk + 4xk
∴ yk = xk+1 – 4xk = (9)k+1 + (2)k+1 – 4(9k + 2k) = 5 . 9k – 2 . 2k
Hence the solution is
xk = {9k + 2k} and yk = {5 . 9k – 2 . 2k}.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. Find the inverse Z-transform of:

z2 + z z2 + 2 z z
(i) 2
(ii) (iii)
( z − 1) ( z − 1)( z − 2)( z − 3) ( z − 1)( z − 2)( z − 3)
z 5z z2
(iv) (v) (vi)
z2 + 11z + 30 (2 z − 1)( z − 3) ( z + 2)( z2 + 4)
4 z −1 9z 3 4 z2 − 2 z
(vii) (viii) F ( z ) = (ix) .
(1 − z −1)2 ( z − 2)(3z − 1)3 z3 − 5 z 2 + 8 z − 4
m
(M.T.U. 2014)
2. Using convolution theorem find the Z–1 of
o
z2 8 z2 z2
(i) (ii) (iii) .
.c
( z − 4)( z − 3) (2 z − 1)(4 z − 1) ( z − a)2
3. Solve the following difference equations:
(i) u(k + 2) – 5u(k + 1) + 6u(k) = 6k if u(0) = u(1) = 0
(ii) yk+2 – 3yk + 1 – 4yk = 0 ; y0 = 3, y1 = – 2
(iii) yk + 2 + yk + 1 – 2yk = 0 ; y0 = 4, y1 = 0
n ts
ra
(iv) yk + 2 – 2yk + 1 + yk = 3k + 5, y(0) = 0, y(1) = 1 (M.T.U. 2014)
by using Z-transform.
4. Solve by Z-transform:
pi
(i) yk+2 – 4yk+1 + 3yk = 5k (ii) yk+2 – 5yk+1 – 6yk = 2k (iii) yk+2 – 6yk+1 + 9yk = 3k.
(G.B.T.U. 2011)
as
(iv) yk +2 – 4yk = k – 1 (v) yk+2 – 6yk+1 + 8yk = 2k + 6k

(vi) yk +1 – 2yk – 1 = 0, k ≥ 1, y(0) = 1 (U.P.T.U. 2015)
g
5. Solve the following difference equations using Z-transform:

.c
(i) yk+2 – 2yk+1 + yk = k ; y0 = y1 = 0 (ii) yk+2 – 4yk = 0 ; y0 = 0, y1 = 2 (U.P.T.U. 2008)

(iii) yk+2 – 2yk+1 + yk = 2 k; y0 = 2, y1 = 1. (iv) yk + 3yk–1 + 2yk–2 = δ(k) + 2δ(k – 1)
w
Find Z– 1
LM 9z 3 OP
w
6. (U.P.T.U. 2015)
MN (3z – 1) ( z – 2) PQ
2
w
Answers
3 5 1 1
1. (i) 2k + 1 (ii) – 4(2)k + (3)k (iii) – (2)k + (3)k (iv) (– 5)k – (– 6)k
2 2 2 2
(v) 3k –
FG 1IJ k
(− 2) k + 1 (2i) k+ 1 (2i) k+ 1
H 2K (vi)
8
+
8i (1 + i)
–
2i (1 − i)
(vii) 2k(k – 1) U(k)
36 F
25k2 + 85k + 72 I FG 1 IJ k
(viii) f(k) =
125
(2) k − GH 250 JK H 3 K
(i) 4k+1 – 3k+1
F 1I – FG 1IJ
(ii) G J
k− 1 2k
(iii) (k + 1) ak U(k)
2.
H 2K H 2K
www.cgaspirants.com
www.cgaspirants.com
1 1 1 1 14
3. (i) u(k) = (6)k – (3)k + (2)k (ii) yk = (4) k + (− 1) k
12 3 4 5 5
8 4 k( k − 1)( k + 3)
(iii) yk = + (− 2) k (iv) yk = +k
3 3 2
5k 2k
4. (i) yk = c1 + c2(3)k + (ii) yk = c1(– 1)k + c2(6)k –
8 12
1 k 1
(iii) yk = (c1 + c2k) (3)k + k(k – 1) (3)k–2 (iv) yk = c1(2)k + c2(– 2)k – +
2 3 9
(v) yk = c1(4)k + c2 −
FG k IJ
(2)k + 2k –
8
(vi) yk =
1
{
( 2 ) k + (− 2 ) k }
H4 K 3 2
k−1
m
5. (i) yk = {1 − (− 1) k } (ii) yk = 2k–1 + (– 2)k–1 (iii) yk = 1 – 2k + 2k
4
(iv) {(– 1)k u(k)}
o
36 FG
5k + 11 IJ FG 1 IJ k
.c
(2) k –
6.
25 25 H K H 3K .
TEST YOUR KNOWLEDGE

n ts
ra
1. Find Fourier cosine and sine transform of f ( x) =

RSk, 0<x<a UV
where k is a constant.
T0, x>a W
pi
2. Find Fourier cosine transform of e–ax cos ax.
3. Find f(x) if its Fourier cosine transform is

1 FG IJ if p < 2a and 0 if p ≥ 2a.
p
as
a−
2π H K
2
z ∞ R1 − α, 0 ≤ α ≤ 1UV
f ( x) sin αx dx = S
g
4. Solve the integral equation

0 T 0, α > 1 W
.c
5. Solve:
z ∞
f ( x) cos αx dx =
RS1 − α,
T 0,
0≤α <1
α >1
w
0
6. Define Fourier transform of a function f(x). (U.P.T.U. 2014)
w
7. Define Fourier sine and cosine transform.

8. Write any two properties of Fourier transform with proof.
w
9. Find the complex Fourier transform of dirac-delta function δ(x – a).

10. State and prove convolution theorem of Fourier transform. (M.T.U. 2014)
11. Define Z-transform of a sequence {f(k)}.
12. Define inverse Z-transform of a function F(z).
13. Define unit step sequence and find its Z-transform.
14. Define unit impulse sequence and find its Z-transform.
15. Find the Z-transform of {nCk}, 0 ≤ k ≤ n. (M.T.U. 2014)
16. State and prove change of scale property of Z-transform.
17. Find the Z-transform of {k3}, k ≥ 0.
18. State and prove convolution theorem for Z-transform.
www.cgaspirants.com
www.cgaspirants.com
R| k2 + 3k + 2 U| , k ≥ 0.
19. Find the Z-transform of S| 2 V|
T W
20. Explain the inverse integral method of finding inverse Z-transform of a function F(Z).
21. Define a difference equation. Define its order and degree. How is Z-transform useful in finding
the solution of a difference equation?
4 z −1
22. Find the inverse Z-transform of .
(1 − z −1)2
23. Solve by Z-transform: yk+2 + yk+1 – 2yk = 0, y0 = 4, y1 = 0.
24. Solve: yk+2 – 5yk+1 – 6yk = 2k by Z-transform.
m
25. Find the Z-transform of {ak}, k ≥ 0. (U.P.T.U. 2014)
Answers
o
k sin ap k(1 − cos ap) a( p2 + 2 a2 ) 2 sin2 ax
.c
1. , 2. 3. .
p p p4 + 4 a4 π2 x2
2 FG x − sin x IJ 2(1 − cos x)
4.
13.
π
z
H x2 K
n 5.
14. 1
πx2
ts 9. eiap
FG
15. 1 + 1 IJ n
z−1 Hz K
ra
17.
z3 + 4 z2 + z
19.
LM
1 z( z + 1)
+
3z
+
2z OP 22. 2k(k – 1) u(k)
( z − 1) 4 MN
2 ( z − 1) 3 ( z − 1) 2 z − 1 PQ
pi
8 4 (2) k z
23. yk = + (– 2)k 24. yk = c1 (– 1)k + c2 (6)k – 25. .
as
3 3 12 z−a
g
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
UNIT 3
Statistical Techniques
m
3.1 MOMENTS
Moments are statistical tools, used in statistical investigations. The moments of a distribu-
o
tion are the arithmetic means of the various powers of the deviations of items from some given
.c
number.
3.2 MOMENTS ABOUT MEAN (Central Moments)
3.2.1. For an Individual Series

n ts
ra
If x1, x2, ..., xn are the values of the variable under consideration, the rth moment μr about
mean x is defined as
pi
∑ (x
i=1
i − x) r
as
μr = ; r = 0, 1, 2, ...
n
3.2.2. For a Frequency Distribution
g
If x1, x2, ..., xn are the values of a variable x with the corresponding frequencies f1, f2, ..., fn
.c
respectively then rth moment μr about the mean x is defined as

w
∑ f (x i i − x)r n
∑f
w
i=1
μr = ; r = 0, 1, 2, ... where N= i
N i=1
w
n n
1 1 N
In particular, μ0 =
N ∑
i=1
f i ( xi − x ) 0 =
N ∑f
i=1
i =
N
=1
∴ For any distribution, μ0 = 1

For r = 1,
1
n
1
n F1 I n
μ1 = ∑ fi ( x i − x ) = ∑ f i xi − x GG N ∑ f JJ = x − x = 0
N i=1
N i=1 H i=1 K
i
∴ For any distribution, μ1 = 0
214
www.cgaspirants.com
www.cgaspirants.com
STATISTICAL TECHNIQUES 215
For r = 2,
n
1
μ2 =
N ∑ f (x
i=1
i i − x ) 2 = (S.D.)2 = Variance
∴ For any distribution, μ2 coincides with the variance of the distribution.

n n
1 1
Similarly, μ3 =
N ∑ fi ( xi − x ) 3 , μ4 =
N ∑ f (x
i=1
i i − x)4
i=1
and so on.
Note. In case of a frequency distribution with class intervals, the values of x are the mid-points of the
intervals.
m
EXAMPLES
o
.c
Example 1. Find the first four moments for the following individual series:
x 3 6 8 10 18
Sol.
n ts
Calculation of Moments
ra
S. No. x x– x (x – x )2 (x – x )3 (x – x )4
1 3 –6 36 – 216 1296
2 6 –3 9 – 27 81
pi
3 8 –1 1 –1 1
4 10 1 1 1 1
as
5 18 9 81 729 6561
n=5 Σx = 45 Σ(x – x ) = 0 Σ(x – x )2 = 128 Σ(x – x )3 = 486 Σ(x – x )4 = 7940

g
.c
Σx 45
Now, x= = =9
n 5
w
Σ( x − x ) 0 Σ( x − x ) 2 128
∴ μ1 = = = 0, μ2 = = = 25.6
n 5 n 5
w
Σ( x − x ) 3 486 Σ( x − x ) 4 7940
μ3 = = = 97.2, μ4 = = = 1588.
w
n 5 n 5
Example 2. Calculate μ1, μ2, μ3, μ4 for the following frequency distribution:
Marks 5–15 15–25 25–35 35–45 45–55 55–65

No. of students 10 20 25 20 15 10
www.cgaspirants.com
www.cgaspirants.com
Sol. Calculation of Moments
Marks No. of Mid- fx x– x f(x – x ) f(x – x )2 f(x – x )3 f(x – x )4

students point
(f) (x)
5–15 10 10 100 – 24 – 240 5760 – 138240 3317760
15–25 20 20 400 – 14 – 280 3920 – 54880 768320
25–35 25 30 750 –4 – 100 400 – 1600 6400
35–45 20 40 800 6 120 720 4320 25920
45–55 15 50 750 16 240 3840 61440 983040
55–65 10 60 600 26 260 6760 175760 4569760
m
N = 100 Σfx Σf(x – x ) Σf(x – x )2 Σf(x – x )3 Σf(x – x )4
= 3400 =0 = 21400 = 46800 = 9671200
o
Σfx 3400
.c
Now, x= = = 34
N 100
Σf ( x − x ) 2 21400
ts
Σf ( x − x ) 0
∴ μ1 = = = 0, μ2 = = = 214
N 100 N 100
n
Σf ( x − x ) 3 46800 Σf ( x − x ) 4 9671200
μ3 = = = 468, μ4 = = = 96712.
N 100 N 100
ra
3.3 SHEPPARD’S CORRECTIONS FOR MOMENTS

pi
While computing moments for frequency distribution with class intervals, we take variable x
as the mid-point of class-intervals which means that we have assumed the frequencies
as
concentrated at the mid-points of class-intervals.

The above assumption is true when the distribution is symmetrical and the no. of class-
g
1 th
intervals is not greater than of the range, otherwise the computation of moments will
20
.c
have certain error called grouping error.

w
This error is corrected by the following formulae given by W.F. Sheppard.

h2
w
μ2 (corrected) = μ2 –
12
1 7 4
w
μ4 (corrected) = μ4 – h 2 μ 2 + h
2 240
where h is the width of the class-interval while μ1 and μ3 require no correction.
These formulae are known as Sheppard’s corrections.
Example 3. Find the corrected values of the following moments using Sheppard’s cor-
rection. The width of classes in the distribution is 10:
μ2 = 214, μ3 = 468, μ4 = 96712.
Sol. We have μ2 = 214, μ3 = 468, μ4 = 96712, h = 10.
h2 (10) 2
Now, μ2 (corrected) = μ2 – = 214 − = 214 – 8.333 = 205.667.
12 12
μ3 (corrected) = μ3 = 468
www.cgaspirants.com
www.cgaspirants.com
1 2 7 4 (10) 2 7
μ4 (corrected) = μ4 – h μ2 + h = 96712 – (214) + (10)4
2 240 2 240
= 96712 – 10700 – 291.667 = 86303.667.
3.4 MOMENTS ABOUT AN ARBITRARY NUMBER (Raw Moments)
If x1, x2, x3, ..., xn are the values of a variable x with the corresponding frequencies f1, f2, f3, ..., fn
respectively then rth moment μr′ about the number x = A is defined as
n n
1
μr′ =
N ∑ fi ( xi − A ) r ; r = 0, 1, 2,... where, N= ∑f
i=1
i
i=1
m
n
1
For r = 0, μ0′ =
N ∑ f (x i i − A )0 = 1
o
i =1
.c
n n n
1 1 A
For r = 1, μ1′ =
N ∑ f (x i i − A) =
N ∑fx i i −
N ∑f i = x –A
For r = 2, μ2′ =
1
N
i=1
nn
∑ f (x
i=1
i i − A) 2
ts
i =1 i =1
ra
n
1
For r = 3, μ3′ =
N ∑ f (x i i − A ) 3 and so on.
pi
i=1
In calculation work, if we find that there is some common factor h (> 1) in values of
x−A
as
x – A, we can ease our calculation work by defining u = . In that case, we have

h
F I
g
n
1
μr′ = GG ∑ f u r
JJ h r
; r = 0, 1, 2, ...
H i i
K
.c
N i=1
Note. For an individual series,

w
n
1
1. μr′ = ∑ ( x − A ) ; r = 0, 1, 2, ...
r
w
i
n i =1
F I
w
n
2. μ ′ =
r
1
G
N G∑
u JJ h ; r = 0, 1, 2, ...
i
r r
for u =
x−A
H K
i=1
h
3.5 MOMENTS ABOUT THE ORIGIN
If x1, x2, ..., xn be the values of a variable x with corresponding frequencies f1, f2, ..., fn respectively
then rth moment about the origin νr is defined as
n n
1
νr =
N ∑ fi xi r ; r = 0, 1, 2,... where, N= ∑f
i=1
i
i=1
www.cgaspirants.com
www.cgaspirants.com
n
1 N
For r = 0, ν0 =
N ∑fx
i =1
i i
0
=
N
=1
n
1
For r = 1, ν1 =
N ∑fx
i=1
i i =x
n
1
For r = 2, ν2 =
N ∑fx
i=1
i i
2
and so on.
3.6 RELATION BETWEEN μr AND μr′
m
We know that,
n
∑ f (x − x)r
o
i i n
1
μr =
i=1
= ∑ f [(x − A) − ( x − A )] r
.c
i i
N N i=1
ts
n
1
=
N ∑ f [( x
i =1
i i − A) − μ ′1 ]r |∵ μ1′ = x – A
n
n
1
∑ f [(x
ra
= i i – A)r – rc1(xi – A)r–1 μ1′ + rc2(xi – A)r–2 μ1′2 – ...... + (– 1)r μ1′r]
N i=1
| Using binomial theorem
pi
⇒ μr = μr′ – rc μ′r–1μ1′ + rc 2 – ...... + (–1)r μ1′r

1 2 μ′r–2μ1′
Putting r = 2, 3, 4, we get
as
μ2 = μ2′ – 2μ1′2 + μ1′2 = μ2′ – μ1′2 |∵ μ0′ = 1

μ3 = μ3′ – 3μ2′μ1′ + 3μ1′3 – μ1′3 = μ3′ – 3μ2′μ1′ + 2μ1′3
g
μ4 = μ4′ – 4μ3′μ1′ + 6μ2′μ1′2 – 3μ1′4

.c
Hence, we have the following relations:

w
μ1 = 0
w
μ2 = μ2′ – μ1′2
w
μ3 = μ3′ – 3μ2′μ1′ + 2μ1′3
and μ4 = μ4′ – 4μ′3μ1′ + 6μ2′μ1′2 – 3μ1′4
3.7 RELATION BETWEEN νr AND μr
We know that,
n
1
νr =
N ∑fx
i=1
i i
r
; r = 0, 1, 2, ...
www.cgaspirants.com
www.cgaspirants.com
n
1
=
N ∑ f (x
i=1
i i − A + A) r
n
1
=
N ∑ f [( x
i=1
i i − A ) r + r c1 ( xi − A ) r − 1 . A + ... + A r ]
= μ r′ + rc
1μr–1 A + ... + A
r
If we take, A = x (for μr) then

νr = μr + rc1μr–1 x + rc2μr–2 x 2 + ... + x r ...(1)
Putting, r = 1, 2, 3, 4 in (1), we get
ν1 = μ1 + μ0 x = x | ∵ μ1 = 0, μ0 = 1
m
ν2 = μ2 + 2c1μ1 x + 2c2 μ0 x 2 = μ2 + x 2
ν3 = μ3 + 3c1μ2 x + 3c2μ1 x 2 + 3c3μ0 x 3 = μ3 + 3μ2 x + x 3
ν 4 = μ4 + 4 c 1 μ 3 x + 4 c 2 μ 2 x 2 + 4 c 3 μ 1 x 3 + 4 c 4 μ 0 x 4
o
= μ4 + 4μ3 x + 6μ2 x 2 + x 4
.c
Hence we have the following relations:
ν1 = x n
ν3 = μ3 + 3μ2 x + x 3
ts
and
ν2 = μ 2 + x 2
ν4 = μ4 + 4μ3 x + 6μ2 x 2 + x 4.
ra
3.8 KARL PEARSON’S β AND γ COEFFICIENTS

pi
Karl Pearson defined the following four coefficients based upon the first four moments of a
frequency distribution about its mean:
as
β1 =
μ 32 U|
μ 32
V| (β-coefficients)
g
μ4
W
.c
β2 =
μ 22
UV
w
γ1 = + β 1 (γ-coefficients)
γ 2 = β2 – 3 W
w
The practical use of these coefficients is to measure the skewness and kurtosis of a
frequency distribution. These coefficients are pure numbers independent of units of measure-
w
ment.
EXAMPLES
Example 1. The first three moments of a distribution, about the value ‘2’ of the variable
are 1, 16 and – 40. Show that the mean is 3, variance is 15 and μ3 = – 86.
Sol. We have A = 2, μ1′ = 1, μ2′ = 16, and μ3′ = – 40
We know that μ1′ = x – A ⇒ x = μ1′ + A = 1 + 2 = 3
Variance = μ2 = μ2′ – μ1′2 = 16 – (1)2 = 15
μ3 = μ3′ – 3μ2′μ1′ + 2μ1′3 = – 40 – 3(16)(1) + 2(1)3 = – 40 – 48 + 2 = – 86.
www.cgaspirants.com
www.cgaspirants.com
Example 2. The first four moments of a distribution, about the value ‘35’ are – 1.8, 240,
– 1020 and 144000. Find the values of μ1 , μ2 , μ3 , μ4 .
Sol. μ1 = 0.
μ2 = μ2′ – μ1′2 = 240 – (– 1.8)2 = 236.76
μ3 = μ3′ – 3μ2′μ1′ + 2μ1′3 = – 1020 – 3(240)(– 1.8) + 2(– 1.8)3 = 264.36
μ4 = μ4′ – 4μ3′μ1′ + 6μ2′μ1′2 – 3μ1′4
= 144000 – 4(– 1020)(– 1.8) + 6(240)(– 1.8)2 – 3(– 1.8)4 = 141290.11.
Example 3. Calculate the variance and third central moment from the following data:
xi 0 1 2 3 4 5 6 7 8
m
fi 1 9 26 59 72 52 29 7 1
(U.P.T.U. 2006)
o
Sol. Calculation of Moments
.c
x− A
x f u= fu fu2 fu3
h
0 1
A = 4, h = 1
n
–4 –4
ts 16 – 64
1 9 –3 – 27 81 – 243
ra
2 26 –2 – 52 104 – 208
3 59 –1 – 59 59 – 59
pi
4 72 0 0 0 0
5 52 1 52 52 52
as
6 29 2 58 116 232
7 7 3 21 63 189
8 1 4 4 16 64
g
N = Σf = 256 Σfu = – 7 Σfu2 = 507 Σfu3 = – 37

.c
Now, moments about the point x = A = 4 are

w
FG Σfu IJ h = − 7 = – 0.02734
μ1′ =
H N K 256
w
F Σfu I h = 507 = 1.9805

2
w
μ′ = G
H N JK 256
2
2
μ′ = G
F Σfu I h = − 37 = – 0.1445
3
H N JK 256
3
3
Moments about mean

μ1 = 0
μ2 = μ2′ – μ′12 = 1.9805 – (– .02734)2 = 1.97975
∴ Variance = 1.97975
Also, μ3 = μ3′ – 3μ2′ μ1′ + 2μ1′3
www.cgaspirants.com
www.cgaspirants.com
= (– 0.1445) – 3 (1.9805) (– .02734) + 2 (– .02734)3

= 0.0178997
∴ Third central moment = 0.0178997.
Example 4. The first three moments of a distribution about the value 2 of the variable
are 1, 16 and – 40 respectively. Find the values of the first three moments about the origin.
Sol. We have A = 2, μ1′ = 1, μ2′ = 16, μ3′ = – 40
∴ ν1 = x = A + μ1′ = 2 + 1 = 3
ν2 = μ2 + x 2 = 15 + (3)2 = 24
ν3 = μ3 + 3μ2 x + x 3 = – 86 + 3(15)(3) + (3)3 = 76.
Example 5. The first four moments of a distribution about x = 2 are 1, 2.5, 5.5 and 16.
m
Calculate the first four moments about the mean and about origin.
Sol. We have A = 2, μ1′ = 1, μ2′ = 2.5, μ3′ = 5.5, μ4′ = 16.
o
Moments about mean
μ1 = 0
.c
μ2 = μ2′ – (μ1′)2 = 2.5 – (1)2 = 1.5
ts
μ3 = μ3′ – 3μ2′μ1′ + 2(μ1′)3 = 5.5 – 3(2.5)(1) + 2(1)3 = 0
μ4 = μ4′ – 4μ3′μ1′ + 6μ2′μ1′2 – 3μ1′4 = 16 – 4(5.5)(1) + 6(2.5)(1)2 – 3(1)4 = 6.
Moments about origin
n
ν1 = x = A + μ1′, ν2 = μ 2 + x 2
ra
ν3 = μ3 + 3μ2 x + x , 3 ν4 = μ4 + 4μ3 x + 6μ2 x 2 + x 4

∴ ν1 = x = 2 + 1 = 3, ν2 = 1.5 + (3)2 = 10.5
pi
3
ν3 = 0 + 3(1.5)(3) + (3) = 40.5, ν4 = 6 + 4(0)(3) + 6(1.5)(3)2 + (3)4 = 168.
Example 6. For a distribution, the mean is 10, variance is 16, γ1 is 1, and β2 is 4. Find
as
the first four moments about the origin.

Sol. x = 10, μ2 = 16, γ1 = 1, β2 = 4 | given
Now, γ1 = 1
g
⇒ β1 = 1 | ∵ γ 1 = β1
.c
μ 23
⇒ =1 ⇒ μ32 = μ23 = (16)3 = (64)2
w
μ 32
⇒ μ3 = 64
w
and β2 = 4
w
μ4
⇒ =4 ⇒ μ4 = 4(16)2 = 1024 |∵ μ2 = 16
μ 22
Moments about the origin
ν1 = x = 10
ν2 = μ2 + x 2 = 16 + 100 = 116
ν3 = μ3 + 3μ2 x + x 3 = 64 + 480 + 1000 = 1544
ν4 = μ4 + 4μ3 x + 6μ2 x 2 + x 4 = 1024 + 4(64)(10) + 6(16)(100) + (10)4
= 22184
Example 7. In a certain distribution, the first four moments about the point x = 4 are
– 1.5, 17, – 30 and 308. Find the moments about mean and about origin. Also, Calculate β1 and β2.
(U.P.T.U. 2014)
www.cgaspirants.com
www.cgaspirants.com
Sol. We have, A = 4, μ′1 = – 1.5, μ′2 = 17, μ′3 = – 30, μ′4 = 308
Moments about mean
μ1 = 0
μ2 = μ′2 – μ1′2 = 17 – (– 1.5)2 = 14.75
μ3 = μ′3 – 3μ′2 μ′1 + 2μ1′3 = – 30 – 3 (17) (– 1.5) + 2 (– 1.5)3 = 39.75
μ4 = μ′4 – 4μ′3 μ′1 + 6μ′2μ1′2 – 3μ1′4
= 308 – 4 (– 30) (– 1.5) + 6(17) (– 1.5)2 – 3 (– 1.5)4 = 342.3125
ν1 = x = μ ′1 + A = – 1.5 + 4 = 2.5
ν2 = μ2 + x 2 = 14.75 + (2.5)2 = 21
m
3
ν3 = μ3 + 3μ2 x + x = 166
o
ν4 = μ4 + 4μ3 x + 6μ2 x 2 + x 4 = 1332
.c
Calculation of β 1 and β2
β1 =
μ 32
μ 32
= 0.492377
n β2 =
μ4
μ 22 ts
= 1.573398
Example 8. The first four moments of a distribution about the value ‘4’ of the variable
ra
are – 1.5, 17, – 30 and 108. Find the moments about mean, about origin ; β1 and β2. Also find
the moments about the point x = 2. (U.P.T.U. 2007)
pi
Sol. We have A = 4, μ′1 = – 1.5, μ′2 = 17, μ′3 = – 30, μ′4 = 108
Moments about mean
as
μ1 = 0
μ2 = μ′2 – μ1′2 = 14.75
μ3 = μ′3 – 3μ′2 μ′1 + 2μ1′3 = 39.75
g
μ4 = μ′4 – 4μ′3 μ′1 + 6μ′2μ1′2 – 3μ1′4 = 142.3125

.c
Also, x = μ′1 + A = – 1.5 + 4 = 2.5

w
ν1 = x = 2.5
ν2 = μ2 + x 2 = 14.75 + (2.5)2 = 21
w
ν3 = μ3 + 3μ2 x + x 3 = 166
w
ν4 = μ4 + 4μ3 x + 6μ2 x 2 + x 4 = 1132

Calculation of β1 and β2
μ 32 μ
β1 = = 0.492377 β2 = 42 = 0.654122
μ 32 μ2
Moments about the point x = 2
μ′1 = x – A = 2.5 – 2 = 0.5
μ′2 = μ2 + μ1′2 = 14.75 + (.5)2 = 15
μ′3 = μ3 + 3μ′2 μ′1 – 2μ1′3 = 39.75 + 3 (15) (.5) – 2 (.5)3 = 62
μ′4 = μ4 + 4μ′3μ′1 – 6μ′2μ1′2 + 3μ1′4 = 244
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. (i) Calculate first four moments about mean, for the following individual series:
5, 5, 5, 5, 5, 5.
(ii) Find the first four moments about the mean of the following series:
1, 3, 7, 9, 10.
(iii) Calculate μ1, μ2, μ3, μ4 for the series : 4, 7, 10, 13, 16, 19, 22.
2. (i) Find the first four moments for the following frequency distribution:
x 1 2 3 4 5 6 7 8 9
f 1 2 3 4 5 4 3 2 1
m
(ii) Calculate the first four moments of the following distribution about the mean and hence find
β1 and β2.
o
x 0 1 2 3 4 5 6 7
.c
f 1 8 28 56 70 56 28 8
(iii) The number of flowers on Sunflower plants are given below:
No. of flowers 3
n
6
ts 12 16 25
ra
No. of plants 1 2 3 4 5
Calculate the first four moments about mean. [M.T.U. (B. Pharma) 2011]
pi
3. (i) Find the first four moments about mean for the following frequency distribution :
Marks 0–10 10–20 20–30 30–40 40–50

as
No. of students 5 10 40 20 25
g
(ii) Calculate the first four moments about the mean for the following:
.c
Classes 5–15 15–25 25–35 35–45 45–55

w
f 14 22 36 18 10
(iii) Calculate the first four moments about the mean for the following data:
w
Class-interval 0–10 10–20 20–30 30–40 40–50

w
f 10 20 40 20 10
(M.T.U. 2014)
4. Calculate the first four moments about x = 15 and hence find the moments about the mean of the
following distribution :
x 10 11 12 13 14 15 16 17 18 19 20 21
f 9 36 75 105 116 107 88 66 45 30 18 5
5. (i) The first three moments of a distribution about the value 4 of the variable are 1.5, 17 and
– 30. Find the moments about mean.
(ii) The first four moments of a distribution about x = 4 are 1, 4, 10 and 45. Show that the mean
is 5, the variance is 3, μ3 is 0 and μ4 is 26.
www.cgaspirants.com
www.cgaspirants.com
(iii) The first four moments of a distribution about the value 5 of the variable are 2, 20, 40 and 50.
Calculate mean, variance, μ3 and μ4. (G.B.T.U. 2011)
6. If the first four moments of a distribution about the value 5 are equal to – 4, 22, – 117 and 560.
Determine the corresponding moments :
(i) about the mean, and (ii) about zero.
7. Compute first four moments of the data 3, 5, 7, 9 about the mean. Also, compute the first four
moments about the point 4.
8. In a frequency distribution, the mean is 1.5, variance is 0.64, β2 is 2.5 and γ1 is 0.3. Find μ3 and μ4
and also the first four moments about the origin.
9. The first four moments of a distribution about the value ‘0’ are – 0.20, 1.76, – 2.36 and 10.88.
Find the moments about the mean. (U.P.T.U. 2009)
Answers
m
1. (i) μ1 = 0, μ2 = 0, μ3 = 0, μ4 = 0 (ii) μ1 = 0, μ2 = 12, μ3 = – 12, μ4 = 208.8
(iii) μ1 = 0, μ2 = 36, μ3 = 0, μ4 = 2268
o
2. (i) μ1 = 0, μ2 = 4, μ3 = 0, μ4 = 37.6 (ii) μ1 = 0, μ2 = 2, μ3 = 0, μ4 = 11, β1 = 0, β2 = 2.75
.c
(iii) μ1 = 0, μ2 = 54.8, μ3 = –49.6, μ4 = 5475.6
3. (i) μ1 = 0, μ2 = 125, μ3 = – 300, μ4 = 37625 (ii) μ1 = 0, μ2 = 134.56, μ3 = 126.14, μ4 = 41840.82
4.
5.
(iii) μ1 = 0, μ2 = 120, μ3 = 0, μ4 = 36000
μ1 = 0, μ2 = 5.5, μ3 = 4.4, μ4 = 77.8
(i) μ1 = 0, μ2 = 14.75, μ3 = – 99.75
n
(iii) 7, 16, – 64, 162
ts
ra
6. (i) 0, 6, 19, 32 (ii) 1, 7, 38, 145
7. μ1 = 0, μ2 = 5, μ3 = 0, μ4 = 41; μ1′ = 2, μ2′ = 9, μ3′ = 38.25, μ4′ = 177
8. μ3 = 0.1536, μ4 = 1.024, ν1 = 1.5, ν2 = 2.89, ν3 = 6.4086, ν4 = 15.6481
pi
9. μ1 = 0, μ2 = 1.72, μ3 = – 1.32, μ4 = 9.4096.

as
3.9 MOMENT GENERATING FUNCTION

[U.P.T.U. 2014 ; G.B.T.U. 2012, M.T.U. 2013]
g
For certain theoretical developments, an indirect method for computing moments is used. The
.c
method depends on the finding of the moment generating function.

3.9.1. In Case of a Continuous Variable x, it is defined as
w
M(t) = z b
e tx f ( x) dx ...(1)
w
a
where integral is a function of parameter t only. The limits a, b can be – ∞ and ∞ respectively.
w
It is possible to associate a moment generating function with the distribution only when all
the moments of the distribution are finite.
Let us see how M(t) generates moments. For this, let us assume that f(x) is a distribu-
tion function for which the integral given by (1) exists.
Then etx may be expanded in a power series and the integration may be performed term
by term. It follows that
M(t) = za
b F 1 + tx + t x
GH
2
2!
2 I
JK
+ ... f(x) dx
= z a
b
f ( x) dx + t z
a
b
xf ( x) dx + ...
www.cgaspirants.com
www.cgaspirants.com
t2
= ν 0 + ν 1 t + ν2 . + ... ...(2)
2!
tr
Obviously, the coefficient of in (2) is the rth moment about the origin.
r!
dr νr
Also, M(t) = r ! + ν r + 1 t + ... = νr ...(3)
dt r t=0
r! t=0
Thus, νr about origin = rth derivative of M(t) with t = 0.

Although the moment generating function (m.g.f.) has been defined for the variable x
only, the definition can be generalized so that it holds for a variable z where z is a function
m
of x. e.g., if z = x – m (m is mean), the rth moment about z will give rth moment of x about the
mean m.
o
Moment generating function for z will clearly be given as
z b
.c
Mz(t) = e tz f ( x) dx
a
Mx–m(t) =
za
b
n e t ( x − m) f ( x) dx = e − mt z
ts b
a
e tx f ( x) dx = e–mt Mx(t).
3.9.2. In Case of Discrete Distribution of the Variable x

ra
We know that, for a variable x,
ν r = Σ xr . P
pi
where P is the probability that the variable takes on the value x.

If z is any function of x, we get rth moment for z by the relation
νr = ΣzrP
as
and the moment generating function is given by

Mz(t) = Σetz P ...(1)
g
tz
To verify that this function generates moments, we will expand e and then sum term
.c
by term,
F I
w
t 2
t2
∴ Mz(t) = ∑ GH1 + tz + 2 ! z 2
JK
+ ... P = ΣP + tΣ zP +
2!
Σ z2 . P + ...
w
t2
= ν0 + tν1 + ν2 + ...
w
2!
dr
In this case, we can also show that νr = M z (t)
dt r t =0
M(t) is clearly the expected value of etx

and hence can be written as E(e tx ) which gives
the moment generating function incase of discrete as well as continuous variables.
Expectation of any function φ(x) is given by
E{φ(x)} = ∑ φ(x ) f ( x )
i
i i | for discrete distribution
or, E{φ(x)} = z ∞
−∞
φ( x) f ( x) dx | for continuous distribution
www.cgaspirants.com
www.cgaspirants.com
Eqn. (1) can also be rewritten as
Mx–a(t) = E [et(x–a)] = ∑e
i
t ( xi − a)
Pi = e–at ∑e txi
Pi = e–at M0(t)
i
Therefore the moment generating function about the point ‘a’ is equal to e–at times the
moment generating function about the origin.
Note. m.g.f. is not always defined since E{| etx |} does not always exist for all values of t.
6
e.g., if f(x) = , x = 1, 2, 3, ... then m.g.f. does not exist.
π2 x2
m.g.f. always exists for t = 0 since Mx=0(0) = 1.
m
3.9.3. Properties of Moment Generating Function (M.T.U. 2013)
(1) The moment generating function of the sum of two independent chance variables is the
product of their respective moment generating functions.
o
Symbolically, Mx+y(t) = Mx(t) × My(t) provided that x and y are independent random
.c
variables.
The m.g.f. of the sum x + y w.r.t. origin is

n ts
Proof. Let x and y be two independent random variables so that x + y is also a random variable.
Mx+y(t) = E{et(x + y)} = E(etx . ety) = E(etx) . E(ety)

ra
Since x and y are independent variables and so are etx and ety.
∴ Mx+y(t) = Mx(t) . My(t)
pi
Hence the theorem.

(2) Effect of change of origin and scale on m.g.f.
as
x-a
Mu(t) = e–at/h Mx (t/h) where u =
h
g
x-a
.c
Proof. Let u be a new random variable given by u = so that x = a + hu

h
w
then by definition, the effect of linear transformation on m.g.f. is governed by

Mx(t) = E(etx) = E[et(a + hu)] = E(eat . ethu)
w
= eat E(ethu) = eat Mu(th)

w
Also, Mu(t) = E(etu)
LM FGH
t
x-a
h
IJ O
KP -
at
FG t IJ
=E e
MN PQ = e h Mx
H hK
(3) Mcx(t) = Mx(ct), c being a constant.
Proof. By definition,
LHS = Mcx (t) = E(etcx) = Mx (ct) = RHS
Hence the result.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Find the moment generating function of the exponential distribution
1
f(x) = e–x/c ; 0 ≤ x ≤ ∞, c > 0 (M.T.U. 2014)
c
Hence find its mean and standard deviation.
Sol. Moment generating function about the origin is given by
Mx(t) = z 0
∞
e tx ⋅
1 −x/c
c
e dx
L FGH IJK OP ∞
1 MM e
1
z
F 1I
m
t− x
∞ Gt − J x
=
1
c
H K
e c dx =
c M F 1I P
c
P
0
MN GH t − c JK PQ
o
0
.c
= (1 – ct)–1 = 1 + ct + c2t2 + c3t3 + ...
LM d M (t)OP
and
∴ ν1 =
ν2 =M
N dt Q
L d M (t)OP
n
2
x
t =0
= 2c2
ts
= (c + 2c2t + 3c3t2 + ...)t= 0 = c
N dt Q 2 x
ra
t =0
Now, mean x = ν1 = c
Variance μ2 = ν2 – x 2 = ν2 – ν12 = 2c2 – c2 = c 2
pi
∴ Standard deviation = μ 2 = c.
as
Example 2. Obtain the moment generating function of the random variable x having
probability distribution
for 0 < x < 1 R| x,
g
f(x) =
for 1 ≤ x < 2 S|2 − x,
.c
elsewhere T 0,
[M.T.U. 2012; G.B.T.U. (C.O.) 2011; G.B.T.U. 2013]
w
Also determine mean ν1, ν2 and variance μ2 .

Sol. Mx(t) = E(etx)
w
= z 1
x . e tx dx + z 2
( 2 − x ) e tx dx + z ∞
0 . e tx dx
w
0 1 2
F xe
=G
I + F 2e − xe + e I
tx
e tx
1 tx tx tx 2
H t JK GH t t t JK
− 2
t 0
2
1
e t
e 1 LF 2 et
2e e I F 2e 2t 2t 2t t
et et I OP = e 2t
− 2e t + 1
+ MG
MNH t t t JK GH t JK PQ
= − + 2
− 2
+ − 2
− + 2
t t t t t t2
F t + t + t + ...I 2 3 2
=G
F e − 1I = GH 2 ! 3 ! JK = 1 + t + t
t 2
H t JK
2 + ...
2
t
www.cgaspirants.com
www.cgaspirants.com
Mean = ν1 =
LM d M (t)OP =1
N dt Q x
t =0
Similarly, ν2 = 2, μ2 = ν2 – x 2 = ν2 – ν12 = 2 – (1)2 = 1 = Variance.

Example 3. Find the moment generating function of the random variable whose moments
are νr = (r + 1) ! 2r.
∞
Sol. Mx(t) = E(etx) = ∑e
x=0
tx
P ( X = x)
∞
tr ∞
tr
∞
= ∑
r=0
r!
νr = ∑ r!
(r + 1) ! . 2r = ∑ (r + 1) (2t) r
r=0 r=0
m
= 1 + 2.2t + 3 . + .... = (1 – (2t)2 2t)–2.
Example 4. Find the moment generating function of the probability distribution function
o
f(z) = e–z (1 + e–z)–2, – ∞ < z < ∞ .
.c
Sol. Mz(t) = E(etz)
= z
−∞
∞
e tz . e − z (1 + e − z ) −2 dz
ts
z ∞
n
= u 2 (u − 1) – t du where 1 + e–z = u ⇒ – e–z dz = du
1
z
ra
1 1 1
= v − t (1 − v) t dv where v = 1 – ⇒ dv = du
0 u u2
pi
= β(1 – t, 1 + t) ; 1 – t > 0
= πt cosec πt, t < 1.
as
Example 5. Find the moment generating function of the negative exponential function
f(x) = λe–λx ; x, λ > 0.
z ∞
z ∞
z ∞
g
Sol. Mx (t) = λ e tx . e − λx dx = λ e (t − λ ) x dx = λ e − (λ − t) x dx
0 0 0
.c
FG IJ ∑ FGH IJK
−1 ∞ r
λ t t
= = 1− =
H ;λ>t
K
w
λ−t λ r=0
λ
Example 6. Find the moment generating function of the discrete binomial distribution
w
given by
P(x) = ncx px qn – x (where q = 1 – p)
w
Also find the first and second moments about the mean. (U.P.T.U. 2008)
Mx(t) = Σ etx . ncx . pxqn – x
= Σ ncx (pet)x qn – x = (q + pet)n
ν1 =
LM d M (t)OP = [n(q + pet)n – 1 . pet ]t = 0 = np |Since q + p = 1
N dt Q x
t=0
ν2 = M
L d M (t)OP 2
N dt Q2 x
t=0
= [np{et . (n – 1) (q + pet)n – 2 pet + (q + pet)n – 1 . et}]t = 0
www.cgaspirants.com
www.cgaspirants.com
= [np (q + pet)n – 2 . et {(n – 1)pet + (q + pet)}]t = 0

= [np(q + pet)n – 2 . et(q + npet)]t = 0
= np(q + np) |∵ q+p=1
= npq + n2 p2
Hence first and second moments about the mean are given by
μ1 = 0
Since x = ν1 = np
∴ μ2 = ν2 – x 2 = ν2 – ν12 = npq + n2p2 – n2p2 = npq
Hence, mean = np, S.D. = μ2 = npq .

Example 7. Find the moment generating function of the discrete Poisson distribution
m
λx
given by P(x) = e–λ ⋅ . Also find the first and second moments about the mean.
o
x!
.c
(M.T.U. 2013)
Mx(t) = Σetx . e –λ .
n λx
x!
= e–λ Σ
(λe t ) x
x! ts t
= e–λ . e λe = e
λ( e t − 1)
LM d M (t)OP t
ra
ν1 = = [eλ(e – 1)
λe t ] t = 0 = λ
N dt Q x
t =0
L d M (t)OP
2
pi
=M
t
λ( e − 1) t
. λet + e
t
ν2 = [λ { e t . e λ( e − 1) e }]
N dt 2
Q x
t =0
t=0
as
= [λe λ ( e − 1) e t (λ e t + 1)]t = 0 = λ(λ + 1)

t
Hence first and second moments about the mean are given by
g
μ1 = 0
.c
Since ν1 = x = λ
∴ μ2 = ν2 – x 2 = ν2 – ν12 = λ (λ + 1) – λ2 = λ
w
Example 8. Find the moment generating function of the continuous normal distribution
w
1
1 − ( x −μ) 2
e 2σ 2
given by f(x) = ;–∞<x<∞.
σ 2π
w
Sol. Moment generating function about the origin is defined as
z
FG
1 x−μ IJ 2
Mx(t) = E (etx) =
∞
e tx ⋅
1
e
−
2 Hσ K dx
−∞ σ 2π
=
e μt
2π z
−∞
∞
e
−
1 2
2
z
. e − tσz dz where z =
x−μ
σ
z
FG μt + 1 t σ IJ2 2 1
=
1
e
H 2 K ∞ − ( z − tσ ) 2
e 2 dz
2π −∞
www.cgaspirants.com
www.cgaspirants.com
= e 2
1
μt + t 2 σ 2
.1= e
1
μt + t 2 σ 2
2
. ∵ z ∞
0
2
e − z dz =
2
π
Example 9. The random variable X assuming only non-negative values has a Gamma
probability distribution if its probability distribution is given by
R| α x β
β −1
e − αx ; x > 0, α > 0, β > 1
U|
f(x) = S Γβ V|
|T0, elsewhere W
Find the moment generating function of Gamma probability distribution.
Sol. Mx(t) = E(etx)
z z
m
∞ α β β − 1 − αx αβ ∞
= e tx ⋅ ⋅x e dx = x β − 1 e − x (α − t) dx
0 Γβ Γβ 0
o
αβ ∞
= yβ − 1 e −β dy | where y = x(α – t) so that dy = (α – t) dx
.c
β
(α − t) Γβ 0
1 1 t FG IJ −β
=
FG 1 − t IJ
H αK
β
n ⋅
Γβ
Γβ = 1 −
α H K
Example 10. Let the random variable X assume the value ‘n’ with the probability law
ts
; | t | < α.
ra
p(X = n) = pqn – 1, n = 1, 2, 3, ... . Find the moment generating function and hence mean and
variance. (G.B.T.U. 2010)
pi
Sol. The given distribution is a discrete distribution.

p t n p t −1 p
Mn(t) = Σ etn pqn – 1 = Σ (e q) = (1 − e q) =
as
q q q (1 − qe t )
which is the moment generating function.
LM d M (t)OP = p LM q e OP = p
g
t
p 1
ν1 = = =
N dt Q q N (1 − q e ) Q (1 − q)
.c
n t 2 2 2
t=0 t=0
p p
L d M (t)OP = p LM d RS e UVOP
w
2 t
ν =M
N dt
2
Q 2
MN dt T (1 − qe ) WPQ
n t 2
w
t=0 t=0
= pM
L (1 − qe ) . e − e . 2 (1 − qe ) (− qe ) OP
t 2 t t t t
w
N (1 − qe ) Q t 4
t=0
= pM
L (1 − q) + 2q (1 − q) OP = 1 + 2q
2
N (1 − q) Q p p 4 2
1
Mean = x = ν1 =
p
1 2q 1 q
Variance = μ2 = ν2 – x 2 = + − =
p p2 p2 p2
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. Define moment generating function. Find the moment generating function of a random variable
X whose probability function is given by:
P(X = x) = p(1 – p)x, x = 0, 1, 2, ..., ∞ (G.B.T.U. 2012)
2. Define moment generating function and two properties of moment generating function with proof.
(M.T.U. 2013)
1
3. The probability density function of the random variable X is f(x) =exp −
|x − θ|
, – ∞ < x < ∞.
FG IJ
2θ θ H K
Find moment generating function of X. Hence find the mean E(X) and variance V(X).
(M.T.U. 2013)
m
LMHint: M (t) = 1
N x
2θ z
−∞
θ
FG
exp −
H
θ − x tx
θ
IJ
e dx +
K
1
2θ z ∞
θ
exp −
FG
H θ
IJ
x − θ tx
e dx
K
OP
Q
o
4. Show that the moment generating function of random variable X having the p.d.f.
.c
1 R| U| R| e 2t
− e −t
f(x) = 3 S|
, − 1< x < 2
V| is MX(t) = S| , t≠0
ts
3t
T
0, elsewhere W T 1 , t=0
5. Find the moment generating function for triangular distribution defined by
n
RS x, 0≤x≤1 UV
ra
f(x) = (M.T.U. 2013)
T2 − x, 1≤ x ≤ 2 W
1
pi
6. If P(X = x) = , x = 1, 2, 3, ..., find the moment generating function of x. Hence obtain the
2x
variance. (U.P.T.U. 2014)
as
Answers
g
p eθt 3θ2t2
1. 3. 2 2
or 1 + θt + + ... ; E(X) = θ, V ( x) = 2θ2
.c
1 − et (1 − p) 1−θ t 2!
1 2 et 1
w
5. M x (t) = 1 + t + t + ... 6. ; .
18 2 − et 2
w
3.10 SKEWNESS
w
For a symmetrical distribution, the frequencies are symmetrically

distributed about the mean i.e., variates equidistant from the
mean have equal frequencies. Also, the mean, mode and median
coincide and median lies half-way between the two quartiles.
Thus, M = M0 = Md and Q 3 – M = M – Q1 . M
M0
Md
Symmetrical distribution
www.cgaspirants.com
www.cgaspirants.com
3.11 MEANING OF SKEWNESS (U.P.T.U. 2015)
If the curve of the distribution is not symmetrical, it may admit of tail on either side of the
distribution. Skewness means lack of symmetry or lopsidedness in a frequency
distribution.
The object of measuring skewness is to estimate the extent to which a distribution is
distorted from a perfectly symmetrical distribution. Skewness indicates whether the curve is
turned more to one side than to other i.e., whether the curve has a longer tail on one side.
Skewness can be positive as well as negative.
o m
.c
3.12 TESTS OF SKEWNESS
n ts
Skewness is positive if the longer tail of the distribution lies towards the right and
negative if it lies towards the left.
ra
1. If A.M. = Mode = Median, then there is no skewness in the distribution. In other

words, the curve of the frequency distribution would be symmetrical, bell-shaped.
pi
2. If A.M. is less than (greater than), the value of mode, the tail would be on left (right)
side, i.e., the distribution is negatively (positively) skewed.
as
3. If sum of frequencies of values less than mode is equal to the sum of frequencies of
values greater than mode, then there would be no skewness.
4. If quartiles are equidistant from median, then there would be no skewness.
g
.c
3.13 METHODS OF MEASURING SKEWNESS (U.P.T.U. 2007)

w
Relative measures of skewness are called the coefficient of skewness. They are independ-
ent of the units of measurement and as such, they are pure numbers.
w
Following are the methods of measuring skewness:

1. Karl Pearson’s Method
w
2. Bowley’s Method
3. Kelly’s Method
4. Method of Moments.
Here, we will discuss Karl Pearson’s method and the method of moments only.
3.13.1. Karl Pearson’s Method
This method is based on the fact that in a symmetrical distribution, the value of A.M. is equal
to that of mode. As we have already noted that the distribution is positively skewed if A.M. >
Mode and negatively skewed if A.M. < Mode. The Karl Pearson’s coefficient of skewness is
given by:
A. M. – Mode
Karl Pearson’s coefficient of skewness =
S. D.
www.cgaspirants.com
www.cgaspirants.com
We have already studied the methods of calculating A.M., Mode and S.D. of frequency
distributions. If mode is ill-defined in some frequency distribution, then the value of empirical
mode is used in the formula.
Empirical mode = 3 Median – 2 A.M.
A.M. − Mode
∴ Coeff. of skewness =
S.D.
A.M. − (3 Median − 2 A.M.) 3 A.M. − 3 Median
= =
S.D. S.D.
3(A. M. – Median)
∴ Karl Pearson’s coefficient of skewness =
S. D.
m
The coefficient of skewness as calculated by using this method gives magnitude as well
as direction of skewness, present in the distribution. Practically, its value lies between – 1 and 1.
o
For a symmetrical distribution, its value comes out to be zero.
.c
The Karl Pearson’s coefficient of skewness is generally denoted by ‘SKP’.
Mean − Mode
(i) If SKP = 0 ⇔ =0
S.D.
⇔ Mean = Mode
n ts
⇔ Distribution is symmetrical.
Thus a distribution is a symmetrical distribution iff SKP = 0.
ra
Mean − Mode
(ii) If SKP > 0 ⇔ >0
S.D.
pi
⇔ Mean – Mode > 0

⇔ Mean > Mode
as
⇔ Distribution is positively skewed.

Thus a distribution is a positively skewed distribution iff SKP > 0.
g
Mean − Mode
(iii) If SKP < 0 ⇔ <0
.c
S.D.
⇔ Mean – Mode < 0
w
⇔ Mean < Mode

⇔ Distribution is negatively skewed.
w
Thus a distribution is a negatively skewed distribution iff SKP < 0.

w
EXAMPLES
Example 1. Karl Pearson’s coefficient of skewness of a distribution is 0.32, its standard
deviation is 6.5 and mean is 29.6. Find the mode of the distribution.
Sol. We have SKP = 0.32, S.D. = 6.5, x = 29.6.
x − Mode
Now SKP =
S.D.
29.6 − Mode
∴ 0.32 =
6.5
⇒ 29.6 – Mode = 0.32 × 6.5 = 2.08
⇒ Mode = 29.6 – 2.08 = 27.52.
www.cgaspirants.com
www.cgaspirants.com
Example 2. For a moderately skewed data, the arithmetic mean is 100, the variance is
35 and Karl Pearson’s coefficient of skewness is 0.2. Find its mode and median.
Sol. We have x = 100, Variance = 35, SKP = 0.2.
x − Mode
Now SKP =
σ
100 − Mode
∴ 0.2 = (∵ S.D. = variance )
35
⇒ 100 – Mode = 0.2 × 5.92 = 1.184
⇒ Mode = 100 – 1.184 = 98.816.
Also, Mode = 3 Median – 2 x ⇒ 98.816 = 3 Median – 2(100).
∴ 3 Median = 98.816 + 200 = 298.816
m
298.816
∴ Median = = 99.61.
3
o
Example 3. In a certain distribution, the following results were obtained :
.c
A.M. = 45, Median = 48, Coefficient of skewness = – 0.4. The person who gave you this
data, failed to give the value of S.D. You are required to estimate it with the help of available
data.
Sol. We have
n ts
coeff. of skewness = – 0.4, A.M. = 45, median = 48.
ra
3( x − Median)
Now, coeff. of skewness =
S.D.
4 3(45 − 48) −9
pi
⇒ – = =
10 S.D. S.D.
⇒ 4 S.D. = 90
as
90
S.D. = = 22.5.
4
g
Example 4. The sum of 20 observations is 300 and sum of their squares is 5000. The
.c
median is 15. Find the Karl Pearson’s coefficient of skewness.

Sol. Let ‘x’ be the variable under consideration.
w
We have n = 20, Σx = 300, Σx2 = 5000, median = 15.

Σx 300
w
Now, x= = = 15
n 20
w
Σx 2 5000
S.D. = − x2 = − (15) 2 = 250 − 225 = 25 = 5
n 20
Now, Karl Pearson’s coeff. of skewness
3( x − Median ) 3(15 − 15) 0

= = = =0.
S.D. 5 5
Example 5. Find the coefficient of skewness by Karl Pearson’s method for the following
data:
Value 6 12 18 24 30 36 42
Frequency 4 7 9 18 15 10 3
www.cgaspirants.com
www.cgaspirants.com
Sol. Calculation of x , S.D.
Value f d=x–A u = d/h fu fu2

x A = 24 h=6
6 4 – 18 –3 – 12 36
12 7 – 12 –2 – 14 28
18 9 –6 –1 –9 9
24 18 0 0 0 0
30 15 6 1 15 15
36 10 12 2 20 40
42 3 18 3 9 27
N = 66 Σfu = 9 Σfu2 = 155
m
FG Σfu IJ h = 24 + FG 9 IJ 6 = 24.82
o
A.M. x = A +
HNK H 66 K
.c
Σfu2
–G
F Σfu IJ × h = 6 × 155 – FG 9 IJ
2 2
Mode.
S.D. =
N
n H NK
Grouping Table
ts 66 H 66 K
= 8.94
ra
x I II III IV V VI
f
pi
6 4
12 7 11
as
18 9 16 20
24 18 27 34
30 15 33 43 42
g
36 10 25 28
.c
42 3 13
w
Analysis Table
Column 24 18 30 36 12
w
I 1
w
II 1 1
III 1 1
IV 1 1 1
V 1 1
VI 1 1 1 1
Total 6 3 3 1 1
∴ Mode = 24
x − Mode 24.82 − 24 0.82
∴ SKP = = = = 0.092.
S.D. 8.94 8.94
www.cgaspirants.com
www.cgaspirants.com
Example 6. Calculate Karl Pearson’s coefficient of skewness for the following data:
Income (in `) 500—600 600—700 700—800 800—900 900—1000 1000—1100

No. of employees 8 12 4 2 1 1
Sol. Calculation of x , Mode, S.D.
Income No. of Mid-points d=x–A u = d/h fu fu2

(in `) employees of classes A = 750 h = 100
f x
500—600 8 550 – 200 –2 – 16 32
m
600—700 12 650 – 100 –1 – 12 12
700—800 4 750 0 0 0 0
o
800—900 2 850 100 1 2 2
.c
900—1000 1 950 200 2 2 4
1000—1100 1 1050 300 3 3 9
N = 28
n ts Σfu = – 21
FG Σfu IJ h = 750 + FG − 21IJ (100) = 750 – 75 = ` 675

Σfu2 = 59
A.M. x=A+
HNK H 28 K
ra
Mode. By inspection, modal class is 600–700

pi
∴ Mode = l +
FG Δ IJ h
1
HΔ +Δ K
as
1 2
Here l = 600, Δ1 = 12 – 8 = 4, Δ2 = 12 – 4 = 8, h = 100
FG 4 IJ (100) = 600 + 33.33 = ` 633.33

g
∴ Mode = 600 +
H 4 + 8K
.c
LM F Σfu IJ OP × h = LM 59 − FG − 21IJ OP × 100

w
2 2
Σfu 2
−G
S.D. =
MN N H N K PQ MN 28 H 28 K PQ
w
= 2.1071 − 0.5625 × 100 = 1.5446 × 100 = 1.2428 × 100 = ` 124.28

w
x − Mode 675 − 633.33

Now, Karl Pearson’s coeff. of skewness = = = 0.34.
S.D. 124.28
ASSIGNMENT
1. A frequency distribution gives the following results:

Coeff. of variation =5
Karl Pearson’s Coeff. of Skewness = 0.5
S.D. =2
Find A.M. and Mode of the distribution.
www.cgaspirants.com
www.cgaspirants.com
2. Find Pearson’s coeff. of skewness from the following frequency distribution:
Height (in inches) 60–62 63–65 66–68 69–71 72–74

Frequency 5 18 42 27 8
3. From the following data, calculate the coefficient of skewness based on mean, median and S.D.
Variable 100–110 110–120 120–130 130–140 140–150 150–160 160–170 170–180

Frequency 4 16 36 52 64 40 32 11
4. From the following data, find out the Karl Pearson’s coefficient of skewness:
m
Measurement 10 11 12 13 14 15
Frequency 2 4 10 8 5 1
o
5. Calculate Karl Pearson’s coefficient of skewness for the following frequency distribution:
.c
Marks more than 0 10 20 30 40 50 60 70
6.
No. of students 100
n 90 75
ts 50 25
For the following frequency distribution, calculate the value of Karl Pearson’s coeff. of skewness:
15 5 0
ra
Temp. (°C) – 40 to – 30 – 30 to – 20 – 20 to – 10 – 10 to 0 0 to 10 10 to 20 20 to 30
No. of days 10 28 30 42 65 180 10

pi
7. From the following data, calculate Karl Pearson’s coefficient of skewness:

as
Marks (above) 0 10 20 30 40 50 60 70 80
No. of students 150 140 100 80 80 70 30 14 0
g
8. Find out the mean wage and coefficient of skewness from the following data:
.c
35 men gets at the rate of ` 4.5 per man

w

w

w

22 men gets at the rate of ` 11.5 per man.
9. Calculate Karl Pearson’s coefficient of skewness:
Wages (in `) 70–80 80–90 90–100 100–110 110–120 120–130 130–140 140–150
No. of workers 12 18 35 42 50 45 20 8
10. Find the mean, mode, S.D. and Karl Pearson’s coeff. of skewness for the following:
Yrs. under 10 20 30 40 50 60
No. of persons 15 32 51 78 97 109
www.cgaspirants.com
www.cgaspirants.com
11. Compute the coeff. of skewness from the following figures : 25, 15, 23, 40, 27, 25, 23, 25, 20.
12. In a discrete series of 20 terms, the sum of the terms is 200, the sum of the squares of the terms
is 5000 and the median is 15. Find Karl Pearson’s coefficient of skewness.
13. Calculate the coefficient of skewness based on mean, median and standard deviation from the
following data:
C.I. 0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80

f 3 6 11 24 28 16 9 3
[M.T.U. (MBA) 2012]

14. Which of the following two series is symmetrical?
Series α: Mean = 22, Median = 24, S.D. = 10
m
Series β: Mean = 22, Median = 25, S.D. = 12
15. Following table gives the data relating to marks obtained by students who appeared for B. Tech.
o
III Semester examination in Mathematics III at a centre. Calculate Karl Pearson’s coefficient of
skewness from the said data:
.c
Marks 0 10 20 30 40 50 60 70
No. of students
to
10
10
n to
20
40
to
30
20
ts
to
40
0
to
50
10
to
60
40
to
70
16
to
80
14
ra
[Hint. Since max. frequency corresponds to two classes very far from each other so mode is ill-
defined and SKP =

3(Mean − Median)
.
OP
Q
pi
S. D.
Answers
as
1. Mean = 40, Mode = 39 2. 0.0356 3. – 0.0087 4. 0.3604

5. – 0.0627 6. – 0.6617 7. – 0.7539
g
8. Mean wage = ` 8.07, Coeff. of skewness = – 0.2422 9. – 0.3314

.c
10. Mean = 29.95, Mode = 35, S.D. = 15.49, Coeff. of skewness = – 0.32 11. – 0.03
12. – 1.22 13. – 0.08608 14. Series α 15. 0.754.
w
3.13.2. Method of Moments

w
In this method, second and third central moments of the distribution are used. This measure
of skewness is called the Moment coefficient of skewness and is given by:
w
μ3
Moment coefficient of skewness = . [G.B.T.U. (C.O.) 2009, 2011]
μ 32
For a symmetrical distribution, its value would come out to be zero. The coefficient of
skewness as calculated by this method gives the magnitude as well as direction of the skewness
present in the distribution.
μ 32
In Statistics, we define β1 = .
μ 32
www.cgaspirants.com
www.cgaspirants.com
μ3
∴ Moment coefficient of skewness can also be written as = = ± β1 .
μ 32
The sign with β 1 is to be taken as that of μ3. The moment coefficient of skewness is
also denoted by γ1. The moment coefficient of skewness is generally denoted by ‘SKM’.
EXAMPLES
Example 1. The first three central moments of a distribution are 0, 15, – 31. Find the
moment coefficient of skewness.
m
Sol. We have μ1 = 0, μ2 = 15 and μ3 = – 31
o
μ3 − 31 31
Moment coefficient of skewness = = =− = – 0.53.
58.09
.c
3 3
μ2 (15)
Example 2. The first four moments of a distribution about the value 5 of the variable
are 2, 20, 40 and 50. Calculate the moment coefficient of skewness.
Sol. We have A = 5, μ1′ = 2, μ2′ = 20, μ3′ = 40
n ts and μ4′ = 50
Now μ2 = μ2′ – (μ1′)2 = 20 – (2)2 = 16
ra
μ3 = μ3′ – 3μ1′μ2′ + 2μ1′3 = 40 – 3(2)(20) + 2(2)3

pi
= 40 – 120 + 16 = – 64
μ3 − 64 − 64
Moment coefficient of skewness = = = = – 1.
as
μ2 3
(16) 3 64
Example 3. Calculate the moment coefficient of skewness for the following distribution :
g
.c
Classes 2.5–7.5 7.5–12.5 12.5–17.5 17.5–22.5 22.5–27.5 27.5–32.5 32.5–37.5
Frequency 8 15 20 32 23 17 5
w
Sol. Calculation of Moment Coefficient of Skewness

w
Classes f Mid-pts. d=x–A u = d/h fu fu2 fu3

w
x A = 20 h=5
2.5–7.5 8 5 – 15 –3 – 24 72 – 216
7.5–12.5 15 10 – 10 –2 – 30 60 – 120
12.5–17.5 20 15 –5 –1 – 20 20 – 20
17.5–22.5 32 20 0 0 0 0 0
22.5–27.5 23 25 5 1 23 23 23
27.5–32.5 17 30 10 2 34 68 136
32.5–37.5 5 35 15 3 15 45 135
N = 120 Σfu = – 2 Σfu2 = 288 Σfu3 = – 62
www.cgaspirants.com
www.cgaspirants.com
FG Σfu IJ h = FG − 2 IJ 5 = – 0.083
Now, μ1′ =
H N K H 120 K
μ′ = G
F Σfu I h = FG 288 IJ 5 = 60
2
H N JK H 120 K
2 2
2
μ′ = G
F Σfu I h = FG − 62 IJ 5 = – 64.583
3
H N JK H 162 K
3 3
3
Now, μ2 = μ2′ – μ1′2 = 60 – (– 0.083)2 = 59.993
m
μ3 = μ3′ – 3μ1′ μ2′ + 2μ1′3 = – 64.583 – 3(– 0.083)(60) + 2(– 0.083)3
= – 49.644.
o
μ3 − 49.644
∴ Moment coefficient of skewness = = = – 0.1068.
.c
μ 32 (59.993) 3
1.
ASSIGNMENT
n ts
The first-three central moments of a distribution are 0, 2.5, 0.7. Find the value of the moment
ra
coefficient of skewness. (U.P.T.U. 2015)
2. In a certain distribution, the first four moments about the point 4 are – 1.5, 17, – 30 and 308.
Calculate the moment coefficient of skewness. (U.P.T.U. 2014)
pi
3. The first three moments of a frequency distribution about origin ‘5’ are – 0.55, 4.46 and – 0.43.
Find the moment coefficient of skewness.
as
4. Calculate the moment coefficient of skewness for the following data:

g
Marks 0–10 10–20 20–30 30–40 40–50 50–60 60–70

.c
No. of students 8 12 20 30 15 10 5
w
5. Calculate the moment coefficient of skewness from the following data:
x 0 1 2 3 4 5 6 7 8
w
f 1 8 28 56 70 56 28 8 1
w
6. For the following frequency distribution, find the first four moments about the mean. Also find
the value of β1. Is it a symmetrical distribution?
x 2 3 4 5 6
f 1 3 7 3 1
7. Compute the coefficient of skewness from the following data: [G.B.T.U. (C.O.) 2009]
x 6 7 8 9 10 11 12
f 3 6 9 13 8 5 4
www.cgaspirants.com
www.cgaspirants.com
8. In two frequency distributions, the second moments about mean are 36 and 49 respectively while
third moments about mean are 43.2 and 85.75. Compare the skewness in the two frequency
distributions. (G.B.T.U. 2012)
n+1 (n + 1)(2n + 1)
9. The first three moments about the origin are given by ν1 = , ν2 = , and
2 6
n(n + 1) 2
ν3 = . Examine the skewness of the data.
4
10. (i) Define skewness of a distribution. (U.P.T.U. 2015)
(ii) Define the coefficients of skewness. (M.T.U. 2012, 2013)
Answers
m
1. 0.17708 2. 0.7017 3. 0.7781 4. 0.0726
o
5. 0 6. 0, 0.933, 0, 2.533, Yes 7. 0.0903.
8. γ1 (for I distribution) = 0.2, γ2 (for II distribution) = 0.25
.c
Second distribution is more positively skewed than the first.
9.
3.14
data is symmetrical.
KURTOSIS
n ts
[U.P.T.U. (C.O.) 2008; U.P.T.U. 2006, 2015]
ra
Given two frequency distributions which have the same variability as measured by the standard
deviation, they may be relatively more or less flat topped than the normal curve. A frequency
curve may be symmetrical but it may not be equally flat topped with the normal curve. The
pi
relative flatness of the top is called kurtosis and is measured by β2. Kurtosis refers to the
bulginess of the curve of a frequency distribution.
as
Curves which are neither flat nor sharply peaked are called normal curves or mesokurtic
curves.
g
Curves which are flatter than the normal curve are called platykurtic curves.
.c
Curves which are more sharply peaked than the normal curve are called leptokurtic
curves.
w
3.15 MEASURE OF KURTOSIS [G.B.T.U. (C.O.) 2009, 2011; U.P.T.U. 2007]

w
The measure of kurtosis is denoted by β2 and is defined as

w
μ4
β2 =
μ 22
where μ2 and μ4 are respectively the second and fourth moments about mean of the distribution.
If β2 > 3, the distribution is leptokurtic. If β2 = 3, the distribution is mesokurtic. If
β2 < 3, the distribution is platykurtic. The kurtosis of a distribution is also measured by
using Greek letter ‘γ2’ which is defined as γ2 = β2 – 3.
∴ γ2 > 0 ⇒ β2 – 3 > 0 ⇒ β2 > 3 ⇒ the distribution is leptokurtic.
Similarly, if γ2 = 0, then β2 = 3 ⇒ The distribution is mesokurtic.
γ2 < 0 ⇒ β2 < 3 ⇒ the distribution is platykurtic.
www.cgaspirants.com
www.cgaspirants.com
Leptokurtic Mesokurtic Platykurtic

b2 > 3 b2 = 3 b2 < 3
g2 > 0 g2 = 0 g2 < 0
3.15.1. Steps for Computing β2

μ4
m
I. If the value of μ2 and μ4 are given, then find β2 by using the formula: β2 = .
μ 22
II. If raw moments μ1′, μ2′, μ3′ and μ4′ are given, then calculate:
o
μ2 = μ2′ – μ1′2 and μ4 = μ4′ – 4μ3′ μ1′ + 6μ2′ μ1′2 – 3μ1′4
.c
μ4
Now, find β2 = .
μ 22
use the formula: β2 =

μ 22
.
n ts
III. If moments are not given, then first find μ2 and μ4 by using the given data and then
μ4
ra
IV. The given distribution is leptokurtic, mesokurtic and platykurtic according as β2 > 3,
β2 = 3 and β2 < 3 respectively.
pi
EXAMPLES
as
Example 1. The first four moments about mean of a frequency distribution are 0, 100,
– 7 and 35000. Discuss the kurtosis of the distribution.
g
Sol. We have, μ1 = 0, μ2 = 100, μ3 = – 7 and μ4 = 35000

.c
μ4 35000
Now, β2 = = = 3.5 > 3
μ 22 (100) 2
w
∴ The distribution is leptokurtic.

w
Example 2. The first four moments of a distribution about the value ‘4’ of the variable
are – 1.5, 17,– 30 and 108. State whether the distribution is leptokurtic or platykurtic.
w
(U.P.T.U. 2007, 2014)

Sol. We have, μ1′ = – 1.5, μ2′ = 17, μ3′ = – 30, μ4′ = 108
Moments about mean:
μ2 = μ2′ – μ1′2 = 17 – (– 1.5)2 = 14.75
μ4 = μ4′ – 4μ3′μ1′ + 6μ2′μ1′2 – 3μ1′4
= 108 – 4 (– 30) (– 1.5) + 6 (17) (– 1.5)2 – 3 (– 1.5)4 = 142.3125
μ4 142.3125
Kurtosis: β2 = = = 0.6541
μ 22 (14.75) 2
Since β2 < 3, the distribution is platykurtic.
www.cgaspirants.com
www.cgaspirants.com
Example 3. The first four moments of a distribution about x = 4 are 1, 4, 10 and 45.
Obtain the various characteristics of the distribution on the basis of the given information.
Comment upon the nature of the distribution.
Sol. We have A = 4, μ1′ = 1, μ2′ = 4, μ3′ = 10 and μ4′ = 45
Moments about mean:
μ1 = 0 (always)
μ2 = μ2′ – μ1′2 = 4 – (1)2 = 3
μ3 = μ3′ – 3μ2′ μ1′ + 2μ1′3 = 10 – 3(4)(1) + 2(1)3 = 0
μ4 = μ4′ – 4μ3′ μ1′ + 6μ2′ μ1′2 – 3μ1′4
= 45 – 4(10)(1) + 6(4)(1)2 – 3(1)4 = 26
m
μ3 0
Skewness: Moment coefficient of skewness, γ1 = = =0.
μ 32 ( 3 )3
o
∴ The distribution is symmetrical.
.c
μ4 26
Kurtosis: β2 = = = 2.89 < 3 ∴ The distribution is platykurtic.
μ 22 (3) 2
ts
Example 4. The standard deviation of a symmetric distribution is 5. What must be the
value of the fourth moment about the mean in order that the distribution be
n
(i) leptokurtic (ii) mesokurtic (iii) platykurtic?
ra
Sol. We have, σ = 5 ⇒ σ2 = 25 ⇒ μ2 = 25 | ∵ μ2 = σ 2
μ μ
Now, β2 = 42 = 4
pi
μ2 625
Thus, the distribution will be
as
μ4
(i) Leptokurtic if β2 > 3 ⇒ > 3 ⇒ μ4 > 1875
625
μ4
g
(ii) Mesokurtic if β2 = 3 ⇒ = 3 ⇒ μ4 = 1875

625
.c
μ4
(iii) Platykurtic if β2 < 3 ⇒ < 3 ⇒ μ4 < 1875.
625
w
Example 5. The first four moments about the working mean 28.5 of a distribution are
0.294, 7.144, 42.409 and 454.98. Calculate the moments about the mean. Also evaluate β1, β2
w
and comment upon the skewness and kurtosis of the distribution. (U.P.T.U. 2006)
w
Sol. We have, μ1′ = 0.294, μ2′ = 7.144, μ3′ = 42.409, μ4′ = 454.98
Moments about mean
μ1 = 0
μ2 = μ2′ – μ1′2 = 7.144 – (.294)2 = 7.0576
μ3 = μ3′ – 3μ2′ μ1′ + 2μ1′3
= 42.409 – 3(7.144) (.294) + 2 (.294)3 = 36.1588
μ4 = μ4′ – 4μ3′ μ1′ + 6μ2′ μ1′2 – 3μ1′4
= 454.98 – 4(42.409) (.294) + 6(7.144) (.294)2 – 3(.294)4
= 408.7896
www.cgaspirants.com
www.cgaspirants.com
Calculation of β 1 and β2
μ 32 μ4
β1 = = 3.7193 β2 = = 8.2070
μ 32 μ 22
Skewness
Since β1 is positive, γ1 = 1.9285 | μ3 is positive
∴ The distribution is positively skewed.
Kurtosis
Since β2 = 8.2070 > 3
∴ The distribution is leptokurtic.
m
Example 6. The first four moments of a distribution about the value ‘0’ are – 0.20, 1.76,
– 2.36 and 10.88. Find the moments about the mean and measure the kurtosis.
(U.P.T.U. 2009)
o
Sol. We have, μ′1 = – 0.20, μ′2 = 1.76, μ′3 = – 2.36, μ′4 = 10.88
.c
Moments about the mean:
μ1 = 0
μ2 = μ′2 – μ1′2 = 1.76 – (–0.20)2 = 1.72
n
μ3 = μ′3 – 3μ′2μ′1 + 2μ1′3
ts
= – 2.36 – 3 (1.76) (– 0.20) + 2 (– 0.20)3 = – 1.32
ra
μ4 = μ′4 – 4μ′3μ′1 + 6μ′2μ1′2 – 3μ1′4

= 10.88 – 4 (– 2.36) (– 0.20) + 6 (1.76) (– 0.20)2 – 3 (– 0.20)4
pi
= 9.4096
Kurtosis::
as
μ4
β2 = = 3.180638
μ 22
g
Since, β2 > 3 hence the distribution is leptokurtic.

.c
Example 7. The following table represents the height of a batch of 100 students. Calcu-
w
late kurtosis.
w
Height (in cm) 59 61 63 65 67 69 71 73 75
No. of students 0 2 6 20 40 20 8 2 2
w
[U.P.T.U. (C.O.) 2008]

Sol. To calculate β2, we will have to first find the values of μ2 and μ4.
Moments about 67
FG Σfu IJ h = FG 12 IJ (2) = 0.24
μ1′ =
H N K H 100 K
μ′ = G
F Σfu I h = FG 164 IJ (4) = 6.56
2
H N JK H 100 K
2
2
www.cgaspirants.com
www.cgaspirants.com
x − 67
Height No. of u= fu fu2 fu3 fu4
2
(cm) x students f
59 0 –4 0 0 0 0
61 2 –3 –6 18 – 54 162
63 6 –2 – 12 24 – 48 96
65 20 –1 – 20 20 – 20 20
67 40 0 0 0 0 0
69 20 1 20 20 20 20
71 8 2 16 32 64 128
73 2 3 6 18 54 162
m
75 2 4 8 32 128 512
o
N = Σf = 100 Σfu = 12 Σfu2 = 164 Σfu3 = 144 Σfu4 = 1100
.c
F Σfu I h = 144 × 8 = 11.52
3
GH N JK 100
μ3′ = 3
μ′ = G
F Σfu I h = 1100 × 16 = 176
n4 ts
H N JK 100
4
4
ra
Moments about mean

μ2 = μ2′ – μ1′2 = 6.56 – (.24)2 = 6.5024
pi
μ4 = μ4′ – 4μ3′ μ1′ + 6μ2′μ1′2 – 3μ1′4

= 176 – 4 (11.52) (0.24) + 6 (6.56) (0.24)2 – 3 (0.24)4 = 167.19798
as
Kurtosis
μ4
Measure of kurtosis β2 = = 3.9544 > 3
g
μ 22
.c
Hence, the distribution is leptokurtic.

w
ASSIGNMENT
w
1. The first four moments about mean of a frequency distribution are 0, 60, – 50 and 8020 respec-
w
tively. Discuss the kurtosis of the distribution.

2. The μ2 and μ4 for a distribution are found to be 2 and 12 respectively. Discuss the kurtosis of the
distribution.
3. (i) The first four central moments of a distribution are 0, 2.5, 0.7 and 18.75. Test the kurtosis of
the distribution. (M.T.U. 2013)
(ii) Define skewness and kurtosis of a distribution. The first four moments of a distribution are
0, 2.5, 0.7 and 18.71. Find the coefficient of skewness and kurtosis. (U.P.T.U. 2015)
4. The standard deviation of symmetric distribution is 4. What must be the value of μ4 so that the
distribution may be mesokurtic?
5. (i) If the first four moments about the value ‘5’ of the variable are – 4, 22, – 117 and 560, find the
value of β2 and discuss the kurtosis.
www.cgaspirants.com
www.cgaspirants.com
(ii) The first four moments of a distribution about the value 5 of the variable are 2, 20, 40 and 50.
Calculate the moments about the mean and comment upon the skewness and kurtosis of the
distribution. (G.B.T.U. 2011)
6. (i) Calculate the value of β2 for the following distribution:
Class 2.5–7.5 7.5–12.5 12.5–17.5 17.5–22.5 22.5–27.5 27.5–32.5 32.5–37.5

Frequency 8 15 20 32 23 17 5
(ii) Compute the value of β2 for the following distribution. Is the distribution platykurtic?
Class 10–20 20–30 30–40 40–50 50–60 60–70 70–80

Frequency 1 20 69 108 78 22 2
m
7. (i) Calculate μ1, μ2, μ3 , μ4 for the frequency distribution of heights of 100 students given in the
following table and hence find coefficient of skewness and kurtosis.
o
.c
Height (cm.) 144.5 – 149.5 – 154.5 – 159.5 – 164.5 – 169.5 – 174.5 –
Class interval 149.5 154.5 159.5 164.5 169.5 174.5 179.5
Frequency 2
n4 13
ts 31 32 15 3
(G.B.T.U. 2011)
ra
(ii) Find all four central moments and discuss skewness and kurtosis for the frequency distribution
given in the following table:
pi
Range of Expenditure 2–4 4–6 6–8 8–10 10–12

(in ` 100 per month)
as
No. of families 38 292 389 212 69
[G.B.T.U. 2013; M.T.U. 2012]

g
8. (i) Find the measures of skewness and kurtosis on the basis of moments for the following
.c
distribution:
x 1 3 5 7 9
w
f 1 4 6 4 1 [G.B.T.U. (C.O.) 2011]

w
(ii) Find the measure of skewness and kurtosis on the basis of moments for the following
distribution:
w
Marks 5–15 15–25 25–35 35–45 45–55

No. of Students 1 3 5 7 4
(M.T.U. (MBA) 2011)

9. Calculate β1 and β2 from the following data:
Profit (in lakhs of ` ) 10–20 20–30 30–40 40–50 50–60

No. of companies 18 20 30 22 10
Indicate the nature of frequency curve.
www.cgaspirants.com
www.cgaspirants.com
10. Prove that the frequency distribution curve of the following frequency distribution is leptokurtic.
Class 10–15 15–20 20–25 25–30 30–35 35–40 40–45 45–50 50–55
Frequency 1 4 8 19 35 20 7 5 1
11. Calculate the first four moments about the mean of the following distribution:
x 2 2.5 3 3.5 4 4.5 5

f 5 38 65 92 70 40 10
Also find the measures of skewness and kurtosis. (M.T.U. 2012)

12. Calculate the first four moments about the mean for the following frequency distribution and
m
hence find the coefficient of skewness and kurtosis and comment upon the nature of the
distribution.
o
Class-interval 5–10 10–15 15–20 20–25 25–30 30–35 35–40
.c
Frequency 6 8 17 21 15 11 2
(G.B.T.U. 2013)
13.
14.
Define the coefficients of kurtosis.
(i) What do you mean by kurtosis? Explain in brief.
(ii) Define kurtosis of a distribution.
n ts
[M.T.U. 2014; G.B.T.U. (C.O.) 2009, 2011; U.P.T.U. 2007]
[U.P.T.U. (C.O.) 2008]
(U.P.T.U. 2006)
ra
Answers
1. β2 = 2.2278, Platykurtic 2. β2 = 3, Mesokurtic
pi
3. (i) β2 = 3, Mesokurtic (ii) 0.17708, 2.9936 4. μ4 = 768

5. (i) β2 = 0.8889, Platykurtic
as
(ii) μ1 = 0, μ2 = 16, μ3 = – 64, μ4 = 162

γ1 = –1, β2 = 0.6328 ; Negatively skewed and platykurtic
g
6. (i) β2 = 2.3216, Platykurtic (ii) β2 = 2.7240, Yes

.c
7. (i) μ1 = 0, μ2 = 36.66, μ3 = – 85.104, μ4 = 4373.3832, γ1 = – 0.3834, β2 = 3.2541

(ii) μ1 = 0, μ2 = 37267.04, μ3 = 1746530.688, μ4 = 3567851989
w
y1 = 0.24275, β2 = 2.5689, positively skewed and platykurtic.

w
8. (i) γ1 = 0, β2 = 2.5
(ii) μ1 = 0, μ2 = 125, μ3 = – 600, μ4 = 37625, γ1 = – 0.4293, β2 = 2.408, negatively skewed and
w
platykurtic.
9. β1 = 0.0001, β2 = 2.047, Platykurtic.
11. μ1 = 0, μ2 = 0.45328125, μ3 = 0.009890625, μ4 = 0.502111743, γ1 = 0.0324, β2 = 2.44379,
positively skewed and platykurtic.
12. μ1 = 0, μ2 = 56, μ3 = – 176.5625, μ4 = 7502.9375, γ1 = – 0.4213, β2 = 2.3925 ; negatively skewed and
platykurtic.
www.cgaspirants.com
www.cgaspirants.com
3.16 CURVE FITTING
Let there be two variables x and y which give us a set of n pairs of numerical values (x1, y1),
(x2, y2).......(xn, yn). In order to have an approximate idea about the relationship of these two
variables, we plot these n paired points on a graph thus, we get a diagram showing the
simultaneous variation in values of both the variables called scatter or dot diagram. From
scatter diagram, we get only an approximate non-mathematical relation between two variables.
Curve fitting means an exact relationship between two variables by algebraic equations, infact
this relationship is the equation of the curve. Therefore, curve fitting means to form an equation
of the curve from the given data. Curve fitting is considered of immense importance both from
the point of view of theoretical and practical statistics.
Theoretically, it is useful in the study of correlation and regression. Practically, it ena-
m
bles us to represent the relationship between two variables by simple algebraic expressions
e.g., polynomials, exponential or logarithmic functions.
o
It is also used to estimate the values of one variable corresponding to the specified
values of the other variable.
.c
The constants occurring in the equation of approximate curve can be found by following
methods:
(i) Graphical method
(iii) Method of least squares
n ts
(ii) Method of group averages
(iv) Method of moments.
Out of the above four methods, we will only discuss and study here method of least
ra
squares.
pi
3.17 METHOD OF LEAST SQUARES

[U.P.T.U. MCA (C.O.) 2008; U.P.T.U. (C.O.) 2008 ; U.P.T.U. 2008]
as
Method of least squares provides a unique set of values to the constants and hence suggests a
curve of best fit to the given data.
g
Suppose we have m-paired observations (x1, y1), (x2, y2), ......, (xm, ym) of two variables x
and y. It is required to fit a polynomial of degree n of the type
.c
y = a + bx + cx2 + ...... + kxn …(1)

w
of these values. We have to determine the constants a, b, c, ..., k such that it represents the
curve of best fit of that degree.
w
In case m = n, we get in general a unique set of values satisfying the given system of
equations.
w
But if m > n then, we get m equations by putting different values of x and y in equa-
tion (1) and we want to find only the values of n constants. Thus, there may be no such solution
to satisfy all m equations.
Therefore, we try to find out those values of a, b, c, ......, k which satisfy all the equations
as nearly as possible. We apply the method of least squares in such cases.
Putting x1, x2, ..., xm for x in (1), we get
y1′ = a + bx1 + cx12 + ...... + kx1n
y2′ = a + bx2 + cx22 + ...... + kx2n
ym′ = a + bxm + cxm2 + ...... + kxmn

where y1′, y2′, ......, ym′ are the expected values of y for x = x1, x2, ......., xm respectively. The values
y1, y2, ......, ym are called observed values of y corresponding to x = x1, x2, ......, xm respectively.
www.cgaspirants.com
www.cgaspirants.com
The expected values are different from observed values, the difference yr – yr′ for differ-
ent values of r are called residuals.
Introduce a new quantity U such that
U = Σ(yr – yr′)2 = Σ(yr – a – bxr – cxr2 – ..... – kxrn)2
The constants a, b, c, ......, k are chosen in such a way that the sum of the squares of
residuals is minimum.
∂U ∂U ∂U ∂U
Now the condition for U to be maximum or minimum is =0= = = ...... = .
∂a ∂b ∂c ∂k
On simplifying these relations, we get
Σy = ma + bΣx + ..... + kΣxn
Σxy = aΣx + bΣx2 + ....... + k Σxn+1
m
Σx2y = aΣx2 + bΣx3 + ....... + k Σxn+2
Σxny = aΣxn + bΣxn+1 + ....... + k Σx2n
o
These are known as Normal equations and can be solved as simultaneous equations to
.c
give the values of the constants a, b, c, ......., k. These equations are (n + 1) in number.
If we calculate the second order partial derivatives and these values are put, they give
a positive value of the function, so U is minimum.
ts
This method does not help us to choose the degree of the curve to be fitted but helps us
n
in finding the values of the constants when the form of the curve has already been chosen.
ra
3.18 FITTING A STRAIGHT LINE
Let (xi, yi), i = 1, 2, ......, n be n sets of observations of related data and

pi
y = a + bx …(1)
be the straight line to be fitted. The residual at x = xi is
as
Ei = yi – f(xi) = yi – a – bxi
g
n n
∑E ∑
.c
2
U= i = ( yi − a − bxi ) 2
i=1 i=1
w
By the principle of Least squares, U is minimum

∂U ∂U
= 0 and =0
w
∴
∂a ∂b
n
∑
w
∴ 2 ( yi − a − bxi )(− 1) = 0 or Σy = na + bΣx ...(2)

i=1
n
and 2 ∑
i=1
( yi − a − bxi )(− xi ) = 0 or Σxy = aΣx + bΣx2 ...(3)
Since xi, yi are known, equations (2) and (3) result two equations in a and b. Solving
these, the best values for a and b can be known and hence equation (1).
Note. In case of change of origin,
x − (middle term)
if n is odd then, u=
interval (h)
x − (mean of two middle terms)
but if n is even then, u= .
1
(interval)
2
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. By the method of least squares, find the straight line that best fits the
following data:
x: 1 2 3 4 5
y: 14 27 40 55 68. (U.P.T.U. 2008)
Sol. Let the straight line of best fit be y = a + bx …(1)
Normal equations are Σy = ma + bΣx …(2)
and Σxy = aΣx + bΣx2 …(3)
Here m = 5
m
Table is as below:
x y xy x2
o
1 14 14 1
.c
2 27 54 4
3 40 120 9
4
5
55
68
n ts
220
340
16
25
Σx = 15 Σy = 204 Σxy = 748 Σx2 = 55

ra
Substituting in (2) and (3), we get

pi
204 = 5a + 15b
748 = 15a + 55b
as
Solving, we get a = 0, b = 13.6
Hence required straight line is y = 13.6x

g
.c
Example 2. Fit a straight line to the following data by least square method:
x: 0 1 2 3 4
w
y: 1 1.8 3.3 4.5 6.3. (U.K.T.U. 2011)

Sol. Let the straight line obtained from the given data be y = a + bx then the normal
w
equations are
Σy = ma + b Σx …(1)
w
Σxy = aΣx + bΣx2 …(2)

Here, m=5
x y xy x2
0 1 0 0
1 1.8 1.8 1
2 3.3 6.6 4
3 4.5 13.5 9
4 6.3 25.2 16
Σx = 10 Σy = 16.9 Σxy = 47.1 Σx2 = 30
www.cgaspirants.com
www.cgaspirants.com
From (1) and (2), 16.9 = 5a + 10b

and 47.1 = 10a + 30b
Solving, we get a = 0.72, b = 1.33
∴ Required line is y = 0.72 + 1.33 x.
Example 3. Show that the best fitting linear function for the points (x1, y1), (x2, y2), .....,
(xn, yn) may be expressed in the form
x y 1
Σx i Σyi n =0 (i = 1, 2, ......, n)
Σx i 2 Σxi yi Σx i
m
Show that the line passes through the mean point ( x , y ) .
Sol. Let the best fitting linear function be y = a + bx …(1)
o
Then the normal equations are
.c
Σyi = na + bΣxi …(2)
and Σxiyi = aΣxi + bΣxi 2 …(3)
Equations (1), (2), (3) may be rewritten as
bx – y + a = 0
n ts
ra
bΣxi – Σyi + na = 0
and bΣxi2 – Σxiyi + aΣxi = 0
pi
Eliminating a and b between these equations

x y 1
as
Σxi Σyi n =0 …(4)

Σxi 2 Σxi yi Σxi
g
which is the required best fitting linear function for the mean point ( x , y ) ,
.c
1 1
x = Σxi, y = Σy .
n n i
w
Clearly, the line (4) passes through point ( x , y ) as two rows of determinant being equal make
w
it zero.
w
ASSIGNMENT
1. Fit a straight line to the following data regarding x as the independent variable:
(i) x 1 2 3 4 5 6
y 1200 900 600 200 110 50
(ii) x 71 68 73 69 67 65 66 67
y 69 72 70 70 68 67 68 64
www.cgaspirants.com
www.cgaspirants.com
(iii) x 0 5 10 15 20 25
y 12 15 17 22 24 30
2. (i) Find the best values of a and b so that y = a + bx fits the given data:
x 0 1 2 3 4
y 1.0 2.9 4.8 6.7 8.6
(ii) Fit a straight line of the form y = a0 + a1x to the data: [U.P.T.U. (C.O.) 2008]
x 1 2 3 4 6 8
y 2.4 3.1 3.5 4.2 5.0 6.0
m
3. Fit a straight line approximate to the data:
o
x 1 2 3 4
.c
y 3 7 13 21
4. A simply supported beam carries a concentrated load P(lb) at its mid-point. Corresponding to
a law of the type Y = a + bP

n ts
various values of P, the maximum deflection Y (in) is measured. The data are given below. Find
P 100 120 140 160 180 200

ra
Y 0.45 0.55 0.60 0.70 0.80 0.85

5. What straight line best fits the following data in the least square sense?
pi
x 1 2 3 4
as
y 0 1 1 2
g
[G.B.T.U. (MCA) 2010]

6. The weight of a calf taken at weekly intervals are given below. Fit a straight line using method
.c
of least squares and calculate the average rate of growth per week.
w
Age 1 2 3 4 5 6 7 8 9 10
Weight 52.5 58.7 65 70.2 75.4 81.1 87.2 95.5 102.2 108.4
w
7. Find the least square line for the data points

w
(– 1, 10), (0, 9), (1, 7), (2, 5), (3, 4), (4, 3), (5, 0) and (6, –1).
8. Find the least square line y = a + bx for the data:
xi –2 –1 0 1 2
yi 1 2 3 3 4
9. If P is the pull required to lift a load W by means of a pulley block, find a linear law of the form
P = mW + c connecting P and W, using the data:
P 12 15 21 25
W 50 70 100 120
where P and W are taken in kg-wt. (U.P.T.U. 2007)
www.cgaspirants.com
www.cgaspirants.com
10. (i) Using the method of least squares, fit a straight line to the following data:
x 1 2 3 4 5
y 2 4 6 8 10
(ii) Using the method of least squares, fit a straight line from the following data:
x 0 2 4 5 6
y 5.012 10 15 21 30 (U.P.T.U. 2009)
(iii) Find the least square line that fits the following data, assuming that x-values are free from
error: [U.P.T.U. MCA (SUM) 2008]
m
x 1 2 3 4 5 6
o
y 5.04 8.12 10.64 13.18 16.20 20.04
.c
Answers
1.
2.
4.
(i) y = 1361.97 – 243.42 x
(i) y = 1 + 1.9x
Y = 0.004P + 0.048
(ii) y = 2.0253 + 0.502 x
n
5. y = – 0.5 + 0.6x
ts
(ii) y = 39.5454 + 0.4242 x (iii) y = 11.285 + 0.7 x
3. y = – 4 + 6x
6. y = 45.74 + 6.16x, 6.16
ra
7. y = – 1.6071429x + 8.6428571 8. y = 2.6 + (0.7) x 9. P = 2.2759 + 0.1879 W
10. (i) y = 2x (ii) y = 3.07734 + 3.86031 x (iii) y = 2.0253 + 2.908 x.
pi
3.19 FITTING OF AN EXPONENTIAL CURVE y = aebx

as
Taking logarithm on both sides, we get

log10 y = log10 a + bx log10 e
g
i.e., Y = A + BX …(1)
.c
where Y = log10 y, A = log10 a, B = b log10 e and X=x

w
The normal equations for (1) are

w
ΣY = nA + BΣX and ΣXY = AΣX + BΣX2

Solving these, we get A and B.
w
B
Then a = antilog A and b = .
log 10 e
3.20 FITTING OF THE CURVE y = axb

log10 y = log10 a + b log10 x
i.e., Y = A + BX …(1)
where Y = log10 y, A = log10 a, B = b and X = log10 x.
www.cgaspirants.com
www.cgaspirants.com
The normal equations to (1) are

ΣY = nA + BΣX
and ΣXY = AΣX + BΣX2
which results A and B on solving and a = antilog A, b = B.
3.21 FITTING OF THE CURVE y = abx

log y = log a + x log b
⇒ Y = A + BX …(1)
m
where Y = log y, A = log a, B = log b, X = x.
o
This is a linear equation in Y and X.
.c
For estimating A and B, normal equations are
ΣY = nA + B ΣX
and ΣXY = A ΣX + B ΣX2
where n is the number of pairs of values of x and y.
n ts
Ultimately, a = antilog (A) and b = antilog (B).
ra
3.22 FITTING OF THE CURVE pvγ = k

pi
pvγ = k ⇒ v = k1/γ p–1/γ

as
1 1
log v = log k − log p
γ γ
g
⇒ Y = A + BX
.c
1 1
w
where Y = log v, A= log k, B = – and X = log p

γ γ
w
γ and k are determined by above equations. Normal equations are obtained as that of the
straight line.
w
3.23 FITTING OF THE CURVE OF TYPE xy = b + ax
b
xy = b + ax ⇒ y=+a
x
1
⇒ Y = bX + a, where X = .
x
Normal equations are ΣY = na + bΣX and ΣXY = aΣX + bΣX2.
www.cgaspirants.com
www.cgaspirants.com
b
3.24 FITTING OF THE CURVE y = ax2 +
x
Let the n points be (x1, y1), (x2, y2), ..... , (xn, yn)
Error of estimate for ith point (xi, yi) is
F b I
GH
Ei = yi − axi −
2
xi
JK
By principle of Least squares, the values of a and b are such that
n
F n
b I 2
U= ∑E
i=1
i
2
= ∑ G y − ax
H i i
2
−
xi
JK is minimum.
i=1
m
Normal equations are given by
n n n
∂U
∑ xi 2 yi = a ∑ xi 4 + b ∑
o
=0 ⇒ xi
∂a i=1 i=1 i=1
.c
n n n
yi 1
and
∂U
=0 ⇒ =a ∑
xi + b ∑ ∑
or
∂b i=1
x i
Dropping the suffix i, normal equations are
n i=1 i=1
xi 2
ts
y 1
∑ ∑
ra
Σx2y = a Σx4 + bΣx and = a Σx + b .

x x2
pi
3.25 FITTING OF THE CURVE y = ax + bx2

CURVE (U.P.T.U. 2014)
as
Error of estimate for ith point (xi, yi) is Ei = (yi – axi– bxi2)
g
n n
U= ∑ Ei 2 = ∑ ( yi − axi − bxi 2 ) 2 is minimum.
.c
i=1 i=1
w
n n n
∂U
=0 ⇒ ∑ xi yi = a ∑ xi 2 + b ∑ xi 3
w
∂a i=1 i=1 i=1

n n n
∂U
∑ ∑ ∑
w
and =0 ⇒ xi 2 yi = a xi 3 + b xi 4
∂b
i=1 i=1 i=1
or Dropping the suffix i, normal equations are
Σxy = a Σx2 + bΣx3 and Σx2y = a Σx3 + bΣx4.
b
3.26 FITTING OF THE CURVE y = ax +
CURVE
x

b
Ei = yi – axi –
xi
www.cgaspirants.com
www.cgaspirants.com

n
Fn
bI
2
U= ∑
i=1
Ei 2 = ∑ GH y − ax − x JK
i=1
i i
i
is minimum.

∂U
=0
∂a
n
F b I
⇒ 2 ∑ GH y
i=1
i − axi −
xi
JK
( − xi ) = 0
n n
⇒ ∑ xi yi = a ∑ xi 2 + nb ...(1)
m
i=1 i=1
∂U
and =0
o
∂b
F I F− 1 I = 0
.c
n
b
⇒ 2 ∑ GH y i − axi −
xi
JK GH x JK
ts
i=1 i
n n
yi 1
∑ ∑
n
⇒ = na + b ...(2)
i=1
xi i=1
xi 2
ra
Dropping the suffix i, normal equations are

y 1
∑ = na + b ∑
pi
Σxy = aΣx2 + nb and

x x2
where n is the no. of pairs of values of x and y.
as
b c
3.27 FITTING OF THE CURVE y = a + + 2
x x
g
.c
Normal equations are

1 1
∑ ∑
w
Σy = ma + b +c
x x2
w
y 1 1 1
∑ x
=a
x∑ +b
x2
+c ∑ ∑ x3
w
y 1 1 1
∑x 2
=a
x
∑2
+b
x3
+c ∑ ∑ x4
where m is number of pairs of values of x and y.
c0
3.28 FITTING OF THE CURVE y = + c1 x
x

c0
E i = yi – − c1 xi
xi
www.cgaspirants.com
www.cgaspirants.com

n n
c0
U= ∑
i=1
Ei2 = ∑
i=1
( yi −
xi
− ci xi ) 2 is minimum.

∂U ∂U
=0 and =0
∂c0 ∂c1
∂U
Now, =0
∂c0
n
F c0 I F− 1 I = 0
m
⇒ 2 ∑ GH y
i=1
i −
xi
− c1 xi JK GH x JK i
o
n n n
yi 1 1
⇒ ∑ = c0 ∑ + c1 ∑ ...(1)
.c
i=1
xi i=1
xi 2 i=1 xi
Also,
n
∂U
∂c1
F
=0
n
c0 I
ts
⇒ 2 ∑ GH y − − c1 xi (− JK xi ) = 0
ra
i
i=1
xi
n n n
pi
1
⇒ ∑
i=1
yi xi = c0 ∑
i=1 xi
+ c1 ∑
i=1
xi ...(2)
as
Dropping suffix i, normal equations (1) and (2) become

y 1 1
∑ = c0 ∑ + c1 ∑
g
2
x x x
.c
1
and ∑ y x = c0 ∑ x
+ c1 ∑ x.
w
3.29 FITTING OF THE CURVE 2x = ax2 + bx + c

w

w
Σ 2xx2 = aΣx4+ bΣx3 + cΣx2

Σ 2x x = aΣx3 + bΣx2 + cΣx
and Σ 2x = aΣx2 + bΣx + mc
where, m is number of points (xi, yi)
3.30 FITTING OF THE CURVE y = ae–3x + be–2x

Σye–3x = a Σe–6x + b Σe–5x
and Σye–2x = a Σe–5x + b Σe–4x
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Find the curve of best fit of the type y = aebx to the following data by the
method of Least squares:
x: 1 5 7 9 12
y: 10 15 12 15 21.
Sol. The curve to be fitted is y = aebx
or Y = A + BX, where, Y = log10 y, A = log10 a, X = x and B = b log10 e
∴ The normal equations are ΣY = 5A + BΣX
and ΣXY = AΣX + BΣX2
m
X=x y Y = log10 y X2 XY
o
1 10 1.0000 1 1
.c
5 15 1.1761 25 5.8805
7 12 1.0792 49 7.5544
ts
9 15 1.1761 81 10.5849
12 21 1.3222 144 15.8664
n
ΣX = 34 ΣY = 5.7536 ΣX2 = 300 ΣXY = 40.8862
ra
Substituting the above values in the normal equations, we get

5.7536 = 5A + 34B
pi
and 40.8862 = 34A + 300B

On solving, A = 0.9766 ; B = 0.02561
as
B
∴ a = antilog10 A = 9.4754 ; b = = 0.059
log 10 e
g
.c
Hence the required curve is y = 9.4754e0.059x.

w
Example 2. Determine the constants a and b by the method of least squares such that
y = aebx fits the following data:
w
x 2 4 6 8 10
w
y 4.077 11.084 30.128 81.897 222.62
Sol. y = aebx
Taking log on both sides
log y = log a + bx log e
or Y = A + BX,
where, Y = log y, A = log a, B = b log10 e, X = x.
ΣY = mA + BΣX ...(1)
and ΣXY = AΣX + BΣX2. ...(2)
Here, m = 5.
www.cgaspirants.com
www.cgaspirants.com
Table is as follows:
x y X Y XY X2
2 4.077 2 .61034 1.22068 4
4 11.084 4 1.04469 4.17876 16
6 30.128 6 1.47897 8.87382 36
8 81.897 8 1.91326 15.30608 64
10 222.62 10 2.347564 23.47564 100
ΣX = 30 ΣY = 7.394824 ΣXY = 53.05498 ΣX2 = 220

Substituting these values in equations (1) and (2), we get
m
7.394824 = 5A + 30B
and 53.05498 = 30A + 220B.
o
Solving, we get A = 0.1760594 and B = 0.2171509
.c
∴ a = antilog (A) = antilog (0.1760594) = 1.49989
B 0.2171509
ts
and b= = = 0.50001
log 10 e 0.4342945
Hence the required equation is
n
y = 1.49989 e0.50001x.
ra
Example 3. Obtain a relation of the form y = abx for the following data by the method of
least squares: [G.B.T.U. MCA (SUM) 2010]
pi
x 2 3 4 5 6
y 8.3 15.4 33.1 65.2 127.4
as
Sol. The curve to be fitted is y = abx

or Y = A + Bx,
g
where, A = log10 a, B = log10 b and Y = log10 y.

.c
∴ The normal equations are ΣY = 5A + BΣx

w
and ΣxY = AΣx + BΣx2.
x y Y = log10 y x2 xY
w
2 8.3 0.9191 4 1.8382

w
3 15.4 1.1872 9 3.5616

4 33.1 1.5198 16 6.0792
5 65.2 1.8142 25 9.0710
6 127.4 2.1052 36 12.6312
Σx = 20 ΣY = 7.5455 Σx2 = 90 ΣxY = 33.1812
Substituting the above values, we get

7.5455 = 5A + 20B and 33.1812 = 20A + 90B.
On solving A = 0.31 and B = 0.3
∴ a = antilog A = 2.04 and b = antilog B = 1.995.
Hence the required curve is y = 2.04(1.995)x.
www.cgaspirants.com
www.cgaspirants.com
Example 4. Obtain the least squares fit of the form f(t) = a e–3t + be–2t for the data:
(U.P.T.U. 2008)
t: 0.1 0.2 0.3 0.4
f(t): 0.76 0.58 0.44 0.35
Sol. Normal equations to the curve f(t) = a e–3t + be–2t are:
Σ f(t) e–3t = a Σ e–6t + b Σ e–5t …(1)
See art. 3.30
Σ f(t) e–2t =aΣ e–5t + bΣ e–4t …(2)
Table of values is
t f(t) e–4t e–5t e–6t f(t) e–2t f(t) e–3t
m
0.1 0.76 0.6703 0.6065 0.5488 0.6222 0.5630
0.2 0.58 0.4493 0.3679 0.3012 0.3888 0.3183
o
0.3 0.44 0.3012 0.2231 0.1653 0.2415 0.1789
.c
0.4 0.35 0.2019 0.1353 0.0907 0.1573 0.1054
Total Σe–4t
= 1.6227
Substituting values in (1) and (2), we get

n Σ e–5t
= 1.3328 ts
Σ e–6t
= 1.106
Σ f(t) e–2t
= 1.4098
Σ f(t) e–3t
= 1.1656
ra
1.106 a + 1.3328 b = 1.1656

1.3328 a + 1.6227 b = 1.4098
pi
On solving, we get a = 0.6778, b = 0.3121.

Hence the least squares fit is f (t) = 0.6778 e–3t + 0.3121e–2t.
as
Example 5. By the method of least squares, find the curve y = ax + bx2 that best fits the
following data: (U.P.T.U. 2014)
g
x 1 2 3 4 5
.c
y 1.8 5.1 8.9 14.1 19.8

w
Sol. Normal equations are

Σxy = aΣx2 + bΣx3 …(1)
w
and Σx2y = aΣx3 + bΣx4 …(2)

w
Let us form a table as below:
x y x2 x3 x4 xy x 2y
1 1.8 1 1 1 1.8 1.8
2 5.1 4 8 16 10.2 20.4
3 8.9 9 27 81 26.7 80.1
4 14.1 16 64 256 56.4 225.6
5 19.8 25 125 625 99 495
Total Σx2 = 55 Σx3 = 225 Σx4 = 979 Σxy = 194.1 Σx2y = 822.9
www.cgaspirants.com
www.cgaspirants.com
Substituting these values in equations (1) and (2), we get

194.1 = 55 a + 225 b
and 822.9 = 225 a + 979 b
83.85 317.4
⇒ ~ 1.52 and b =
a= ~ .49
55 664
Hence required parabolic curve is y = 1.52 x + 0.49 x2.
Example 6. Fit the curve pvγ = k to the following data: [U.P.T.U. MCA (C.O.) 2007]
p (kg/cm2) 0.5 1 1.5 2 2.5 3
v (litres) 1620 1000 750 620 520 460
m
Sol. pvγ = k
FG k IJ
o
1/ γ
⇒ v= = k1/γ p–1/γ
H pK
.c
1 1
Taking log,
which is of the form

log v =
γ
log k − log p
n
Y = A + BX
γ
ts
1 1
ra
where Y = log v, X = log p, A = log k and B=–

γ γ
p v X Y XY X2
pi
0.5 1620 – 0.30103 3.20952 – 0.96616 0.09062

1 1000 0 3 0 0
as
1.5 750 0.17609 2.87506 0.50627 0.03101

2 620 0.30103 2.79239 0.84059 0.09062
g
2.5 520 0.39794 2.716 1.08080 0.15836

.c
3 460 0.47712 2.66276 1.27046 0.22764
Total ΣX = 1.05115 ΣY = 17.25573 ΣXY = 2.73196 ΣX2 = 0.59825

w
Here, m = 6
w

17.25573 = 6A + 1.05115 B
w
and 2.73196 = 1.05115 A + 0.59825 B

Solving these, we get
A = 2.99911 and B = – 0.70298
1 1
∴ γ=– = = 1.42252
B 0.70298
Again, log k = γA = 4.26629
∴ k = antilog (4.26629) = 18462.48
Hence, required curve is
pv1.42252 = 18462.48.
www.cgaspirants.com
www.cgaspirants.com
Example 7. The pressure of the gas corresponding to various volumes V is measured,

given by the following data:
V (cm3): 50 60 70 90 100
P (kg cm–2): 64.7 51.3 40.5 25.9 78
Fit the data to the equation PV = C. γ
Sol. PV γ = C
⇒ P = CV– γ
Taking log on both sides, we get
log P = log C – γ log V
m
⇒ Y = A + BX
where, Y = log P, A = log C, B = – γ, X = log V
o
.c
ΣY = mA + BΣX …(1)
and
Here m = 5
ΣXY = AΣX +
n BΣX2
ts …(2)
The table is as below:

ra
V P X = log V Y = log P XY X2
pi
50 64.7 1.69897 1.81090 3.07666 2.88650
60 51.3 1.77815
as
1.71012 3.04085 3.16182
70 40.5 1.84510 1.60746 2.96592 3.40439

g
90 25.9 1.95424 1.41330 2.76193 3.81905

.c
100 78 2 1.89209 3.78418 4

w
ΣX = 9.27646 ΣY = 8.43387 ΣXY = 15.62954 ΣX2 = 17.27176

w
From Normal equations, we have

w
8.43387 = 5A + 9.27646 B
and 15.62954 = 9.27646 A + 17.27176 B
A = 2.22476, B = – 0.28997
∴ γ = – B = 0.28997
C = antilog (A) = antilog (2.22476) = 167.78765
Hence, the required equation of curve is
PV0.28997 = 167.78765.
www.cgaspirants.com
www.cgaspirants.com
Example 8. (i) Given the following experimental values:

x: 0 1 2 3
y: 2 4 10 15
Fit by the method of Least squares a parabola of the type y = a + bx2.
(ii) Find the Least squares fit of the form y = a0 + a1x2 to the following data:
x: –1 0 1 2
y: 2 5 3 0 (U.P.T.U. 2008)
Sol. (i) Error of estimate for ith point (xi, yi) is Ei = (yi – a – bxi 2)
By method of Least squares, the values of a, b are chosen such that
m
4 4
U= ∑ Ei 2 = ∑ ( yi − a − bxi 2 ) 2 is minimum.
o
i=1 i=1
Normal equation are given by
.c
∂U
=0 ⇒ Σy = ma + bΣx2 …(1)
ts
∂a
∂U
and =0 ⇒ Σx2y = aΣx2 + bΣx4 …(2)
n
∂b
ra
x y x2 x 2y x4
0 2 0 0 0
pi
1 4 1 4 1
2 10 4 40 16
as
3 15 9 135 81
Total Σy = 31 Σx2 = 14 Σx2y = 179 Σx4 = 98

g
.c
Here m=4
From (1) and (2), 31 = 4a + 14b and 179 = 14a + 98b
w
Solving for a and b, we get a = 2.71, b = 1.44

Hence the required curve is y = 2.71 + 1.44 x2.
w
(ii) Normal equations are

w
Σy = ma0 + a1 Σx2 ...(1)

and Σx2y = a0 Σx2 + a1 Σx4 ...(2)
The table is as follows:
x y x2 x 2y x4
–1 2 1 2 1
0 5 0 0 0
1 3 1 3 1
2 0 4 0 16
Σy = 10 Σx2 = 6 Σx2y = 5 Σ x4 = 18
www.cgaspirants.com
www.cgaspirants.com
Here, m=4
From (1) and (2), 10 = 4a0 + 6a1
5 = 6a0 + 18a1
⇒ a0 = 4.1667, a1 = – 1.1111
Hence, the required curve is y = 4.1667 – 1.1111 x2
c0
Example 9. Use the method of Least squares to fit the curve: y = + c1 x to the following
x
table of values: [G.B.T.U. (MCA) 2007, 2011]
x: 0.1 0.2 0.4 0.5 1 2
m
y: 21 11 7 6 5 6.
c0
o
Sol. As derived in art. 3.28, normal equations to the curve y = + c1 x are
x
.c
y 1 1
∑ x
= c0 ∑ x 2
+ c1 ∑ x
...(1)
and ∑ y x = c0
n
∑
1
x
+ c1 ts
∑x ...(2)
Table is as below:
ra
1 1
x y y/x
pi
y x x x2
0.1 21 210 6.64078 3.16228 100

as
0.2 11 55 4.91935 2.23607 25

0.4 7 17.5 4.42719 1.58114 6.25
g
0.5 6 12 4.24264 1.41421 4

.c
1 5 5 5 1 1
w
2 6 3 8.48528 0.70711 0.25
1
∑ ∑x
1
w
Σx = 4.2 Σ(y/x) = 302.5 Σy x = 33.71524 = 10.10081 2

= 136.5
x
w
From equations (1) and (2), we have

302.5 = 136.5 c0 + 10.10081 c1
and 33.71524 = 10.10081 c0 + 4.2 c1
c0 = 1.97327 and c1 = 3.28182
Hence the required equation of curve is
1.97327
y= + 3.28182 x .
x
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. (i) Using the method of least squares, fit the non-linear curve of the form y = aebx to the following
data:
x 0 2 4
y 5.012 10 31.62
(ii) The voltage V across a capacitor at time t seconds is given by the following table. Use the
principle of least squares to fit a curve of the form V = aekt to the data:
t 0 2 4 6 8
m
V 150 63 28 12 5.6
(iii) For the data given below, find the equation to the best fitting exponential curve of the form
o
y = aebx.
.c
x 1 2 3 4 5 6
2.
y 1.6 n
(i) State some important curve-fitting procedures.
4.5 13.8
ts 40.2 125 300
(U.P.T.U. 2008)
b
ra
(ii) Derive the least square equations for fitting a curve of the type y = ax + to a set of n points
x
(xi, yi) ; i = 1, 2, …, n.
pi
3. (i) Fit a curve y = axb to the following data:
x 1 2 3 4 5 6
as
y 2.98 4.26 5.21 6.1 6.8 7.5

g
(ii) Fit a least square geometric curve y = axb to the following data:
.c
x 1 2 3 4 5
w
y 0.5 2 4.5 8 12.5

w
(iii) Fit a curve of the form y = axb to the data given below:
x 1 2 3 4 5
w
y 7.1 27.8 62.1 110 161
4. (i) Fit a curve of the form y = abx in least square sence to the data given below:
x 2 3 4 5 6
y 144 172.8 207.4 248.8 298.5
(ii) Fit an exponential curve of the form y = abx to the following data:
x 1 2 3 4 5 6 7 8
y 1 1.2 1.8 2.5 3.6 4.7 6.6 9.1
www.cgaspirants.com
www.cgaspirants.com
5. The pressure and volume of a gas are related by the equation pva = b where a and b are constants.
Fit this equation to the following set of data:
p (kg/cm2) 0.5 1 1.5 2 2.5 3

v (litres) 1.62 1 0.75 0.62 0.52 0.46
6. (i) Determine the constants of the curve y = ax + bx2 for the following data:
x 0 1 2 3 4
y 2.1 2.4 2.6 2.7 3.4
(ii) Using method of least squares, derive the normal equations to fit the curve y = ax2 + bx.
m
Hence fit this curve to the following data: (G.B.T.U. 2011)
x 1 2 3 4 5 6 7 8
o
y 1 1.2 1.8 2.5 3.6 4.7 6.6 9.1
.c
7. Fit a curve of the type xy = ax + b to the following data:
x
y
1
36
3
n
29
5
28
ts 26
7 9
24
10
15
ra
b
8. Fit a relation y = ax + which satisfies the following data, using method of least squares:
x
pi
x 1 2 3 4 5 6 7 8
as
y 5.4 6.2 8.2 10.3 12.6 14.8 17.2 19.5
(G.B.T.U. 2010)
g
b
9. Derive the least square equations for fitting a curve of the type y = ax2 + to a set of n points.
.c
x
Hence fit a curve of this type to the data:
w
x 1 2 3 4
w
y – 1.51 0.99 3.88 7.66

w
10. Derive the least squares approximations of the type ax2 + bx + c to the function 2x at the points
xi = 0, 1, 2, 3, 4.
11. A person runs the same race track for 5 consecutive days and is timed as follows:
Day (x) 1 2 3 4 5
Time (y) 15.3 15.1 15 14.5 14
b c
Make a least square fit to the above data using a function a + + .
x x2
www.cgaspirants.com
www.cgaspirants.com
c1
12. Use the method of least squares to fit the curve y = c0 x + for the following data:
x
x 0.2 0.3 0.5 1 2

y 16 14 11 6 3
13. Experiments with a periodic process gave the following data:
t° 0 50 100 150 200 250 300 350
y 0.754 1.762 2.041 1.412 0.303 – 0.484 – 0.38 0.520
m
Estimate the parameters a and b in the model y = b + a sin t using the least squares approximation.
14. A physicist wants to approximate the following data:
o
x 0 0.5 1 2
.c
f(x) 0 0.57 1.46 5.05
15. ts
using a function a ebx + c. He believes that b ≈ 1. Compute the values of a and c that give the best
least squares approximation assuming that b = 1.
Determine the normal equations if the cubic polynomial y = a0 + a1x + a2x2 + a3x3 is fitted to the
data (xi, yi ); 0 ≤ i ≤ m.
n
ra
b
16. Estimate y at x = 5 by fitting a least squares curve of the form y = to the following data:
x ( x − a)
pi
x 3.6 4.8 6 7.2 8.4 9.6 10.8

as
y 0.83 0.31 0.17 0.10 0.07 0.05 0.04
FG Hint: Rewrite the equation as 1 a 1 IJ

g
= − x + x2
H y b b K
.c
Answers
w
1. (i) y = 4.642 e0.46x (ii) V = 146.3 e–0.4118t (iii) y = 0.5580 e1.0631 x

3. (i) y = 2.978 x0.5143 (ii) y = 0.5012 x1.9977 (iii) y = 7.173x1.952
w
4. (i) y = 99.86 (1.2)x (ii) y = 0.6823 (1.384)x 5. pv1.42 = 0.99

w
6. (i) a = 1.97, b = – 0.298 (ii) y = 0.107798 x2 + 0.217125 x 7. xy = 16.18x + 40.78
2.98195 2.04
8. y = 2.39188 x + 9. y = 0.509x2 –
x x
6.7512 4.4738
10. y = 1.143x2 – 0.971x + 1.286 11. y = 13.0065 + −
x x2
7.5961
12. y = – 1.1836 x + 13. a = 1.312810, b = 0.752575
x
14. a = 0.784976, b = – 0.733298
www.cgaspirants.com
www.cgaspirants.com
15. Σy = m a 0 + a 1 Σx + a 2 Σx 2 + a 3 Σx 3
Σxy = a 0 Σx + a 1 Σx 2 + a 2 Σx 3 + a 3 Σ x 4
Σx 2 y = a 0 Σx 2 + a 1 Σx 3 + a 2 Σx 4 + a 3 Σx 5
Σx 3 y = a 0 Σx 3 + a 1 Σx 4 + a 2 Σx 5 + a 3 Σx 6 .
3.774
16. y= ; y (5) = 0.2516
x ( x − 2)
3.31 CORRELATION
In a bivariate distribution, if the change in one variable affects a change in the other variable,
the variables are said to be correlated.
m
If the two variables deviate in the same direction i.e., if the increase (or decrease) in one
results in a corresponding increase (or decrease) in the other, correlation is said to be direct or
o
positive.
e.g., the correlation between income and expenditure is positive.
.c
If the two variables deviate in opposite direction i.e., if the increase (or decrease) in one
or negative.
n ts
results in a corresponding decrease (or increase) in the other, correlation is said to be inverse
e.g., the correlation between volume and the pressure of a perfect gas or the correlation
between price and demand is negative.
ra
Correlation is said to be perfect if the deviation in one variable is followed by a corre-

sponding proportional deviation in the other.
pi
3.32 REASONS RESPONSIBLE FOR THE EXISTENCE OF CORRELATION

CORRELATION
as
1. Due to mere chance. The correlation between variables may be due to mere chance.
Consider the data regarding six students selected at random from a college.
g
Students: A B C D E F
.c
% of marks obtained in: 43% 47% 60% 80% 55% 40%

previous exam.
w
Height (in inches): 60 62 65 70 64 59

Here the variables are moving in the same direction and a high degree of correlation is
w
expected between the variables. We cannot expect this degree of correlation to hold good for
any other sample drawn from the concerned population. In this case, the correlation has oc-
w
curred just due to chance.

2. Due to the effect of some common cause. The correlation between variables may be due
to the effect of some common cause. For example, positive correlation between the number of
girls seeking admission in colleges A and B of a city may be due to the effect of increasing
interest of girls towards higher education.
3. Due to the presence of cause-effect relationship between variables. For example, a
high degree correlation between ‘temperature’ and ‘sale of coffee’ is due to the fact that people
like taking coffee in winter season.
4. Due to the presence of interdependent relationship between the variables. For
example, the presence of correlation between amount spent on entertainment of family and
total expenditure of family is due to fact that both variables affect each other.
www.cgaspirants.com
www.cgaspirants.com
3.33 SCATTER OR DOT DIAGRAMS
It is the simplest method of the diagrammatic representation of bivariate data. Let (xi, yi),
i = 1, 2, 3,......, n be a bivariate distribution. Let the values of the variables x and y be plotted
along the x-axis and y-axis on a suitable scale. Then corresponding to every ordered pair,
there corresponds a point or dot in the xy-plane. The diagram of dots so obtained is called a dot
or scatter diagram.
If the dots are very close to each other and the number of observations is not very large,
a fairly good correlation is expected. If the dots are widely scattered, a poor correlation is
expected.
3.34 KARL PEARSON’S CO-EFFICIENT OF CORRELATION (OR PRODUCT
m
MOMENT CORRELATION CO-EFFICIENT)
[U.P.T.U. (C.O.) 2009; U.P.T.U. 2006, 2007 , 2015]
o
Correlation co-efficient between two variables x and y, usually denoted by r(x, y) or rxy is a
.c
numerical measure of linear relationship between them and is defined as
1 1
rxy =
Σ ( xi − x )( yi − y)
Σ ( xi − x ) 2 Σ ( yi − y ) 2
n =
1
n
n
2 1 ts
Σ ( xi − x )( yi − y)
Σ ( xi − x ) . Σ ( yi − y)
n
2
= n
Σ ( xi − x )( yi − y )
σ xσ y
.
ra
Σ( x − x )( y − y)
∴ rxy =
nσ x σ y
pi
Alternate form of r(x, y):

as
n Σxy − ΣxΣy
r(x, y) =
n Σx − ( Σx ) 2
2
n Σy 2 − (Σy) 2
g
.c
Here n is the no. of pairs of values of x and y.

Note. Correlation co-efficient is independent of change of origin and scale.
w
Let us define two new variables u and v as

x−a y−b
u= ,v= where a, b, h, k are constants, then rxy = ruv.
w
h k
w
n Σuv − ΣuΣv
Then, r(u, v) = .
n Σu − (Σu) 2
2
n Σv2 − (Σv) 2
EXAMPLES
Example 1. Find the coefficient of correlation between the values of x and y:
[U.P.T.U. (C.O.) 2008]
x 1 3 5 7 8 10
y 8 12 15 17 18 20
www.cgaspirants.com
www.cgaspirants.com
Sol. Here, n = 6. The table is as follows:

x y x2 y2 xy
1 8 1 64 8
3 12 9 144 36
5 15 25 225 75
7 17 49 289 119
8 18 64 324 144
10 20 100 400 200
Σ x = 34 Σ y = 90 Σ x2 = 248 Σ y2 = 1446 Σ xy = 582
m
Karl Pearson’s coefficient of correlation is given by
n Σ xy − Σ x Σ y
r (x, y) =
o
n Σ x 2 − ( Σ x) 2 n Σ y 2 − (Σ y) 2
.c
(6 × 582) − (34 × 90)
= = 0.9879
(6 × 248) − (34) 2
ts
(6 × 1446) − (90) 2
Example 2. The following data regarding the heights (y) and weights (x) of 100 college
students are given:
n
ra
Σx = 15000, Σx2 = 2272500, Σy = 6800, Σy2 = 463025 and Σxy = 1022250.
Find the correlation coefficient between height and weight.
Sol. Here, n = 100
pi
Correlation co-efficient r(x, y) is given by

as
nΣ xy − Σ xΣ y
r=
nΣ x 2 − ( Σ x ) 2 n Σy 2 − ( Σy ) 2
g
(100 × 1022250) − (15000 × 6800)

.c
= = 0.6.
(100 × 2272500) − (15000) 2 (100 × 463025) − (6800) 2
w
Example 3. Find the co-efficient of correlation for the following table: (U.K.T.U. 2011)
x: 10 14 18 22 26 30
w
y: 18 12 24 6 30 36
w
x − 22 y − 24
Sol. Let u = ,v=
4 6
x y u v u2 v2 uv
10 18 –3 –1 9 1 3
14 12 –2 –2 4 4 4
18 24 –1 0 1 0 0
22 6 0 –3 0 9 0
26 30 1 1 1 1 1
30 36 2 2 4 4 4
Total Σu = – 3 Σv = – 3 Σu2 = 19 Σv2 = 19 Σuv = 12
www.cgaspirants.com
www.cgaspirants.com
1 1 1 1 1 1
Here, n = 6, u= Σu = (− 3) = − ; v = Σv = (− 3) = −
n 6 2 n 6 2
n Σuv − ΣuΣv
ruv =
n Σu − (Σu) 2
2
n Σv2 − (Σv) 2
(6 × 12) − (− 3)(− 3) 63
= = = 0.6
(6 × 19) − (− 3) 2
(6 × 19) − (− 3) 2 105 105
Hence, rxy = ruv = 0.6.
Example 4. Ten students got the following percentage of marks in Principles of Economics
and Statistics:
m
Roll Nos.: 1 2 3 4 5 6 7 8 9 10
Marks in Economics: 78 36 98 25 75 82 90 62 65 39
o
Marks in Statistics: 84 51 91 60 68 62 86 58 53 47
Calculate the co-efficient of correlation.
.c
Sol. Let the marks in the two subjects be denoted by x and y respectively.
x
78
36
84
51
y u = x – 65
13
– 29
n v = y – 66
18
– 15
ts 169
841
u2 v2
324
225
uv
234
435
ra
98 91 33 25 1089 625 825
25 60 – 40 –6 1600 36 240
75 68 10 2 100 4 20
pi
82 62 17 –4 289 16 – 68
90 86 25 20 625 400 500
as
62 58 –3 –8 9 64 24
65 53 0 – 13 0 169 0
39 47 – 26 – 19 676 361 494
g
Total Σu = 0 Σv = 0 Σu2 = 5398 Σv2 = 2224 Σuv = 2734

.c
1 1
Here, n = 10, u = Σui = 0, v = Σvi = 0
n
w
n
nΣ uv − Σ uΣ v
ruv =
w
n Σu 2 − ( Σu ) 2 n Σv 2 − ( Σv) 2
(10 × 2734) − (0 × 0)
w
= = 0.789
(10 × 5398) − (0) 2 (10 × 2224) − (0) 2
Hence, rxy = ruv = 0.789.
Example 5. A computer while calculating correlation co-efficient between two variables
X and Y from 25 pairs of observations obtained the following results :
n = 25, ΣX = 125, ΣX2 = 650,
ΣY = 100, 2
ΣY = 460, ΣXY = 508.
It was, however, later discovered at the time of checking that he had copied down two
pairs as X Y while the correct values were X Y
6 14 8 12
8 6 6 8
Obtain the correct value of correlation co-efficient.
www.cgaspirants.com
www.cgaspirants.com
Sol. Corrected ΣX = 125 – 6 – 8 + 8 + 6 = 125 U|

Corrected
Corrected
ΣY = 100 – 14 – 6 + 12 + 8 = 100
ΣX2 = 650 – 62 – 82 + 82 + 62 = 650
|V
Corrected ΣY2 = 460 – 142 – 62 + 122 + 82 = 436 ||
Corrected ΣXY = 508 – 6 × 14 – 8 × 6 + 8 × 12 + 6 × 8 = 520 W
(Subtract the incorrect values and add the corresponding correct values)
1 1 1 1
X= ΣX = × 125 = 5 ; Y= ΣY = × 100 = 4
n 25 n 25
n ΣXY − ΣXΣY
Corrected rxy =
n ΣX − ( ΣX ) 2
2
n ΣY 2 − ( ΣY ) 2
m
(25 × 520) − (125 × 100)
= = 0.67.
o
(25 × 650) − (125) 2 (25 × 436) − (100) 2
.c
Example 6. If z = ax + by and r is the correlation co-efficient between x and y, show that
Sol.
⇒
σz2 = a2σx2 + b2σy2 + 2abr σx σy.
z = ax + by
z = ax + by ,
n
zi = axi + byi
ts
ra
zi – z = a( xi − x ) + b( yi − y )
1 1
Now, σz2 = Σ( zi − z ) 2 = Σ[a(xi – x ) + b(yi – y )]2
pi
n n
1
= Σ[a2(xi – x )2 + b2(yi – y )2 + 2ab(xi – x )(yi – y )]
as
n
1 1 1
= a2 . Σ ( xi − x ) 2 + b2 . Σ ( yi − y) 2 + 2ab . Σ ( xi − x )( yi − y)
n n n
g
1
Σ( xi − x )( yi − y)
.c
2 2 2 2
= a σx + b σy + 2abr σxσy ∵ r= n
σ xσ y
w
Example 7. Establish the formula: σ 2x − y = σx2 + σy2 – 2rxyσxσy

w
where rxy is the correlation coefficient between x and y. Using the above formula, calculate the
w
correlation coefficient from the following data relating to the marks of 10 candidates in aptitude
test (x) and Achievement rating (y).
Marks
Aptitude (x): 22 53 46 67 43 35 88 11 95 13
Achievement (y): 18 39 31 42 55 64 82 10 96 14
Sol. Let z=x–y
∴ z=x−y
∴ z – z = ( x – x ) − ( y − y)
or, ( z − z ) 2 = ( x − x ) 2 + ( y − y) 2 − 2( x − x )( y − y)
www.cgaspirants.com
www.cgaspirants.com
Summing up for n terms,

Σ(z – z )2 = Σ(x – x )2 + Σ(y – y )2 – 2Σ(x – x )(y – y )
Σ ( z − z ) 2 Σ ( x − x ) 2 Σ ( y − y) 2 2Σ ( x − x )( y − y)
or, = + −
n n n n
Σ( x − x )( y − y )
⇒ σz2 = σx2 + σy2 – 2rσxσy, where r =
nσ x σ y
⇒ σ 2x − y = σx2 + σy2 – 2rσxσy ...(1)
Now, n = 10
m
22 + 53 + 46 + 67 + 43 + 35 + 88 + 11 + 95 + 13 473
x= = = 47.3
10 10
o
18 + 39 + 31 + 42 + 55 + 64 + 82 + 10 + 96 + 14 451
y= = = 45.1
.c
10 10
Now we form the table as
x
22
y
18
z=x–y
4 – 25.3
x– x
ny– y
– 27.1 1.8
ts
z– z (x – x )2
640.09
(y – y )2
734.41
(z – z )2
3.24
ra
53 39 14 5.7 – 6.1 11.8 32.49 37.21 139.24
46 31 15 – 1.3 – 14.1 12.8 1.69 198.81 163.84
67 42 25 19.7 – 3.1 22.8 38.09 9.61 519.84
pi
43 55 – 12 – 4.3 9.9 – 14.2 18.49 98.01 201.64

35 64 – 29 – 12.3 18.9 – 31.2 151.29 357.21 973.44
as
88 82 6 40.7 36.9 3.8 1656.49 1361.61 14.44

11 10 1 – 36.3 – 35.1 – 1.2 1317.69 1232.01 1.44
95 96 –1 47.7 50.9 – 3.2 2275.29 2590.81 10.24
g
13 14 –1 – 34.3 – 31.1 – 3.2 1176.49 967.21 10.24

.c
Σ(x – x )2 Σ(y – y )2 Σ(z – z )2

w
= 7658.1 = 7586.9 = 2037.6
Σz 22
w
where, z= = = 2.2
n 10
w
Σ( x − x ) 2 7658.1 Σ( y − y) 2 7586.9
Now, σx2 = = = 765.81, σy2 = = = 758.69
n 10 n 10
Σ( z − z ) 2 2037.6
σ 2x − y = σ z 2 = = = 203.76.
n 10
Substituting the values in the formula (1),
σ 2x − y = σx2 + σy2 – 2r σxσy
⇒ 203.76 = 765.81 + 758.69 – 2r (27.67)(27.54)
1524.5 − 203.76
r= = 0.866.
1524.06
www.cgaspirants.com
www.cgaspirants.com
Example 8. (i) Calculate coefficient of correlation from the following results:

n = 10, ΣX = 100, ΣY = 150, Σ(X – 10)2 = 180, Σ(Y – 15)2 = 215, Σ(X – 10)(Y – 15) = 60.
(ii) Calculate Karl Pearson’s coefficient of correlation between X and Y for the following
information:
n = 12, ΣX = 120, ΣY = 130, Σ(X – 8)2 = 150, Σ(Y – 10)2 = 200 and Σ(X – 8)(Y – 10) = 50
ΣX 100
Sol. (i) Mean of first series, X= = = 10
n 10
ΣY 150
Mean of second series, Y = = = 15
n 10
m
Σ (X − X)(Y − Y)
Now, r=
Σ (X − X) 2 . Σ (Y − Y) 2
o
Σ (X − 10)(Y − 15) 60
= = = 0.305
.c
2
Σ (X − 10) . Σ (Y − 15) 2 180 × 215
(ii) Σx = Σ(X – 8) = ΣX – Σ8 = 120 – (8 × 12) = 24

Σy = Σ(Y – 10) = ΣY – Σ10 = 130 – (10 × 12) = 10
n
Σxy = Σ(X – 8)(Y – 10) = 50 (given)
ts
ra
Σx2 = Σ(X – 8)2 = 150
Σy2 = Σ(Y – 10)2 = 200
pi
n Σxy − ΣxΣy (12 × 50) − (24 × 10)

Now, r= =
n Σx 2 − ( Σx ) 2 n Σy 2 − (Σy) 2 (12 × 150) − (24) 2 (12 × 200) − (10) 2
as
360
= = 0.2146.
g
1224 2300
.c
3.35 CALCULATION OF CO-EFFICIENT OF CORRELATION FOR A BIVARIATE

FREQUENCY DISTRIBUTION
w
If the bivariate data on x and y is presented on a two way correlation table and f is the frequency
w
of a particular rectangle in the correlation table, then

w
1
Σfxy − Σfx Σfy
rxy = n
LMΣfx 2
−
1
(Σfx) 2
OP LMΣfy 2
−
1
(Σfy) 2
OP
N n QN n Q
Since change of origin and scale do not affect the co-efficient of correlation.
∴ rxy = ruv where the new variables u, v are properly chosen.
Example 9. The following table gives according to age the frequency of marks obtained
by 100 students in an intelligence test:
www.cgaspirants.com
www.cgaspirants.com
Age (in years)

18 19 20 21 Total
Marks
10–20 4 2 2 8
20–30 5 4 6 4 19
30–40 6 8 10 11 35
40–50 4 4 6 8 22
50–60 2 4 4 10
60–70 2 3 1 6
m
Total 19 22 31 28 100
Calculate the co-efficient of correlation between age and intelligence.
o
Sol. Let age and intelligence be denoted by x and y respectively.
.c
Mid x
value
15 10–20
y
18
4
19
2
n 20
2
21
tsf
8
u
–3
fu
– 24
fu2
72
fuv
30
ra
25 20–30 5 4 6 4 19 –2 – 38 76 20
35 30–40 6 8 10 11 35 –1 – 35 35 9
pi
45 40–50 4 4 6 8 22 0 0 0 0
55 50–60 2 4 4 10 1 10 10 2
as
65 60–70 2 3 1 6 2 12 24 –2
f 19 22 31 28 100 Total – 75 217 59
g
v –2 –1 0 1 Total
.c
fv – 38 – 22 0 28 – 32
w
fv2 76 22 0 28 126
fuv 56 16 0 – 13 59
w
y − 45
w
Let us define two new variables u and v as u = , v = x – 20

10
1
Σfuv − Σfu Σfv
rxy = ruv = n
LMΣfu 2
−
1
(Σfu) 2
OP LMΣfv 2
−
1
(Σfv) 2
OP
N n QN n Q
1
59 − (− 75)(− 32)
100 59 − 24
= = = 0.25.
LM 217 −
1
(− 75) 2
126 −
1OP LM
(− 32) 2 OP 643 2894
×
N 100 100QN Q 4 25
www.cgaspirants.com
www.cgaspirants.com
3.36 RANK CORRELATION
Sometimes we have to deal with problems in which data cannot be quantitatively measured
but qualitative assessment is possible.
Let a group of n individuals be arranged in order of merit or proficiency in possession of
two characteristics A and B. The ranks in the two characteristics are, in general, different.
For example, if A stands for intelligence and B for beauty, it is not necessary that the most
intelligent individual may be the most beautiful and vica versa. Thus an individual who is
ranked at the top for the characteristic A may be ranked at the bottom for the characteristic
B. Let (xi, yi), i = 1, 2, ...... n be the ranks of the n individuals in the group for the characteristics
A and B respectively. Pearsonian co-efficient of correlation between the ranks xi’s and yi’s is
called the rank correlation co-efficient between the characteristics A and B for that group of
m
individuals.
Thus rank correlation co-efficient
o
1
Σ ( xi − x )( yi − y )
Σ ( xi − x )( yi − y )
= n
.c
r= ...(1)
Σ ( x − x ) 2 Σ ( y − y) 2 σ xσ y
i i
ts
Now xi’s and yi’s are merely the permutations of n numbers from 1 to n. Assuming that
no two individuals are bracketed or tied in either classification i.e., (xi , yi) ≠ (xj , yj) for i ≠ j, both
n
x and y take all integral values from 1 to n.
1 1 n(n + 1) n + 1
ra
∴ x = y = (1 + 2 + 3 + ...... + n) = . =
n n 2 2
n(n + 1)
pi
Σxi = 1 + 2 + 3 + ...... + n = = Σyi

2
n(n + 1)(2n + 1)
Σxi2 = 12 + 22 + ...... + n2 = = Σyi 2
as
6
If Di denotes the difference in ranks of the ith individual, then
g
Di = xi – yi = (xi – x ) – (yi – y ) [∵ x = y ]
1 1
.c
ΣD i 2 = Σ[( xi − x ) − ( yi − y)]2
n n
w
1 2 1 2 1
= Σ ( xi − x ) + Σ ( yi − y) − 2 . Σ ( xi − x )( yi − y)
n n n
= σx2 + σy2 – 2rσxσy
w
...(2) | using (1)

1 1
But σ x2 = Σxi2 – x 2 = Σyi2 – y 2 = σy2
w
n n
∴ From (2),
1
ΣDi2 = 2σx2 – 2rσx2 = 2(1 – r) σx2 = 2(1 – r)
1
Σxi 2 − x 2
LM OP
n n N Q
= 2(1 – r)
LM 1 . n(n + 1)(2n + 1) − (n + 1) OP
2
Nn 6 4 Q
= (1 – r)(n + 1) M
L 4n + 2 − 3n − 3 OP = (1 − r)(n 2
− 1)
or 1–r=
6 ΣD i 2
N 6 Q 6 n(n 2 − 1)
Hence, r=1– M
L 6ΣD OP i
2
MN n(n − 1) PQ
2
www.cgaspirants.com
www.cgaspirants.com
Note. This is called Spearman’s Formula for Rank Correlation.

Σdi = Σ(xi – yi) = Σxi – Σyi = 0 always.
This serves as a check on calculations.
EXAMPLES
Example 1. Compute the rank correlation coefficient for the following data:
Person: A B C D E F G H I J
Rank in Maths: 9 10 6 5 7 2 4 8 1 3
Rank in Physics: 1 2 3 4 5 6 7 8 9 10
Sol. Here the ranks are given and n = 10.
m
Person R1 R2 D = R1 – R2 D2
o
A 9 1 8 64
B 10 2 8 64
.c
C 6 3 3 9
ts
D 5 4 1 1
E 7 5 2 4
n
F 2 6 –4 16
G 4 7 –3 9
ra
H 8 8 0 0
I 1 9 –8 64
pi
J 3 10 –7 49
ΣD2 = 280
as
∴ r=1–
RS 6ΣD UV = 1 − RS 6 × 280 UV = 1 – 1.697 = – 0.697.
2
g
T n(n − 1) W T 10(100 − 1) W
2
.c
Example 2. The marks secured by recruits in the selection test (X) and in the proficiency
test (Y) are given below:
w
Serial No.: 1 2 3 4 5 6 7 8 9
X: 10 15 12 17 13 16 24 14 22
w
Y: 30 42 45 46 33 34 40 35 39
Calculate the rank correlation co-efficient.
w
Sol. Here the marks are given. Therefore, first of all, write down ranks. In each series,
the item with the largest size is ranked 1, next largest 2 and so on. Here n = 9.
X 10 15 12 17 13 16 24 14 22 Total
Y 30 42 45 46 33 34 40 35 39
Ranks in X (x) 9 5 8 3 7 4 1 6 2
Ranks in Y (y) 9 3 2 1 8 7 4 6 5
D=x–y 0 2 6 2 –1 –3 –3 0 –3 0
D2 0 4 36 4 1 9 9 0 9 72
www.cgaspirants.com
www.cgaspirants.com
∴ r=1–
RS 6ΣD UV = 1 − RS 6 × 72 UV = 1 – .6 = 0.4
2
T n(n − 1) W T 9 × 80 W
2
Example 3. Rank correlation co-efficient of marks obtained by 10 students in Mathematics

and English was found to be 0.5. It was later discovered that the difference in ranks in two
subjects obtained by one of the students was wrongly taken as 3 instead of 7. Find the correct
rank correlation co-efficient.
Sol. Incorrect rk = 0.5, n = 10
n(n 2 − 1)(1 − rk ) 10 × 99 × 0.5
∴ Incorrect ΣD2 = = = 82.5
6 6
Now, correct ΣD2 = Incorrect ΣD2 – (3)2 + (7)2 = 82.5 – 9 + 49 = 122.5
m
∴ Correct rk = 1 –
RS 6ΣD UV = 1 − RS 6 × 122.5 UV = 0.2575.
2
T n(n − 1) W T 10(100 − 1) W
2
o
Example 4. Ten competitors in a beauty contest were ranked by three judges in the
.c
following orders:
First Judge:
Second Judge:
Third Judge:
1
3
6
6
5
4
n 5
8
9
10
4
8
ts
3
7
1
10
2
2
4
2
3
9
1
10
7
6
5
8
9
7
ra
Use the method of rank correlation to determine which pair of judges has the nearest
approach to common taste in beauty?
pi
Sol. Let R1, R2, R3 be the ranks given by three judges.

Calculation of rank correlation coefficient
as
Competitor R1 R2 R3 D12 D13 D23 2

D12 2
D13 2
D23
= R1 – R 2 = R1 – R3 = R2 – R3
g
A 1 3 6 –2 –5 –3 4 25 9
.c
B 6 5 4 1 2 1 1 4 1
C 5 8 9 –3 –4 –1 9 16 1
w
D 10 4 8 6 2 –4 36 4 16
E 3 7 1 –4 2 6 16 4 36
w
F 2 10 2 –8 0 8 64 0 64
G 4 2 3 2 1 –1 4 1 1
w
H 9 1 10 8 –1 –9 64 1 81
I 7 6 5 1 2 1 1 4 1
J 8 9 7 –1 1 2 1 1 4
n = 10 ΣD122 ΣD132 2
ΣD23
= 200 = 60 = 214
Rank correlation coefficient between first and second judges,

R|S 6ΣD U|V = 1 – R 6 × 200 U = – 0.212
2
rk12 = 1 – 12
T| n(n − 1) W|
2 ST 10 (99) VW
www.cgaspirants.com
www.cgaspirants.com
Rank correlation coefficient between first and third judges,

R|S 6ΣD U|V = 1 – R 6 × 60 U = 0.636
2
rk13 = 1 –
|T n(n − 1) |W
2
13
ST 10 × 99 VW
Rank correlation coefficient between second and third judges,
r = 1 – |S
R 6ΣD U|V = 1 – R 6 × 214 U = – 0.297
2
k23
T| n(n − 1) W|
2 ST 10 × 99 VW
23
Correlation between first and second judges is negative i.e., their opinions regarding
beauty test are opposite to each other. Similarly, opinions of second and third judges are
opposite to each other, but the opinions of first and third judges are of similar type as their
correlation is positive. It means that their likings and dislikings are very much common.
m
3.37 TIED RANKS
o
If any two or more individuals have same rank or the same value in the series of marks, then
.c
the above formula fails and requires an adjustment. In such cases, each individual is given an
average rank. This common average rank is the average of the ranks which these individuals
ts
would have assumed if they were slightly different from each other. Thus, if two individual
are ranked equal at the sixth place, they would have assumed the 6th and 7th ranks if they
n
6+7
were ranked slightly different. Their common rank = = 6.5. If three individuals are
ra
2
ranked equal at fourth place, they would have assumed the 4th, 5th and 6th ranks if they
4+5+6
pi
were ranked slightly different. Their common rank = = 5.

3
1
as
Adjustment. Add m(m2 – 1) to ΣD2 where m stands for the number of times an item is
12
repeated.
g
This adjustment factor is to be added for each repeated item.

RS UV
.c
1 1
6 ΣD 2 + m(m 2 − 1) + m(m 2 − 1) + ...
T 12 12 W
w
Thus r=1– .
n(n 2 − 1)
w
Example 5. Obtain the rank correlation co-efficient for the following data:
X: 68 64 75 50 64 80 75 40 55 64
w
Y: 62 58 68 45 81 60 68 48 50 70.
Sol. Here, marks are given, so write down the ranks.
X 68 64 75 50 64 80 75 40 55 64 Total
Y 62 58 68 45 81 60 68 48 50 70
Ranks in X (x) 4 6 2.5 9 6 1 2.5 10 8 6
Ranks in Y (y) 5 7 3.5 10 1 6 3.5 9 8 2
D=x–y –1 –1 –1 –1 5 –5 –1 1 0 4 0
D2 1 1 1 1 25 25 1 1 0 16 72
www.cgaspirants.com
www.cgaspirants.com
In the X-series, the value 75 occurs twice. Had these values been slightly different, they
2+3
would have been given the ranks 2 and 3. Therefore, the common rank given to them is
2
= 2.5. The value 64 occurs thrice. Had these values been slightly different, they would have
5+6+7
been given the ranks 5, 6, and 7. Therefore the common rank given to them is = 6.
3
Similarly, in the Y-series, the value 68 occurs twice. Had these values been slightly different,
they would have been given the ranks 3 and 4? Therefore, the common rank given to them is
3+4
= 3.5.
2
Thus, m has the values 2, 3, 2.
m
RS 1
6 ΣD 2 + m (m − 1) +
1 2
m (m − 1) +
1
m (m 2 2
− 1)
UV
T 12
1
12
1
12
2 2 3 3
W
o
∴ r=1– 2
n(n − 1)
.c
L
6 M72 +
1
. 2(2 − 1) +2 1
. 3(3 − 1) +
1 O
. 2(2 − 1)P
2 2
N 12 12 12 Q
=1–
R 6 × 75 UV = 6 = 0.545.
=1– S
n 10(10 − 1)
ts
2
T 990 W 11
ra
ASSIGNMENT
pi
1. Calculate the coefficient of correlation for the following data: (U.P.T.U. 2006)
as
Husband’s age (in yrs.) x 23 27 28 28 29 30 31 33 35 36

g
Wife’s age (in yrs.) y 18 20 22 27 21 29 27 29 28 29

.c
2. Calculate the coefficient of correlation for the following data:

w
Height of father 65 66 67 67 68 69 70 72
(in inches)
w
Height of son 67 68 65 68 72 72 69 71
(in inches)
w
3. Define Karl Pearson’s coefficient of correlation. How would you interpret the sign and magnitude
of a correlation coefficient?
4. Calculate the coefficient of correlation between the marks obtained by 8 students in Mathematics
and Statistics:
Students A B C D E F G H
Mathematics 25 30 32 35 37 40 42 45
Statistics 08 10 15 17 20 23 24 25
[U.P.T.U. (C.O.) 2009]
www.cgaspirants.com
www.cgaspirants.com
5. Find the correlation coefficient between x and y for the following data:
x 60 34 40 50 45 41 22 43
y 75 32 34 40 45 33 12 30
[U.K.T.U. 2010]
6. The marks of the same 15 students in two subjects A and B are analysed. The two numbers
within the brackets denote the ranks of the same student in A and B respectively:
(1, 10) (2, 7) (3, 2) (4, 6) (5, 4) (6, 8) (7, 3) (8, 1) (9, 11) (10, 15) (11, 9) (12, 5) (13, 14) (14, 12)
(15, 13)
Use Spearman’s formula to find the rank correlation coefficient.
7. Ten students got the following percentage of marks in Chemistry and Physics:
m
Students: 1 2 3 4 5 6 7 8 9 10
Marks in Chemistry: 78 36 98 25 75 82 90 62 65 39
o
Marks in Physics: 84 51 91 60 68 62 86 58 63 47
Calculate the rank correlation co-efficient.
.c
8. A firm not sure of the response to its product in ten different colour shades decides to produce
them in those colour shades, if the ranking of these colour shades by two typical consumer judges
is highly correlated.
The two judges rank the ten colours in the following order:
n ts
Colour : Red Green Blue Yellow White Black Pink Purple Orange Ivory
ra
Ranking
by I : 6 4 3 1 2 7 9 8 10 5
pi
Judge
Ranking
as
by II : 4 1 6 7 5 8 10 9 3 2
Judge
Is there any agreement between the two judges, to allow the introduction of the product by the
g
firm in the market?

.c
9. Two judges in a music competition rank the 12 entries as follows:
x 1 2 3 4 5 6 7 8 9 10 11 12
w
y 12 9 6 10 3 5 4 7 8 2 11 1
w
What degree of agreement is there between the judgement of the two judges?
10. Calculate the coefficient of correlation between the following ages of husband (x) and wife (y) by
w
taking 30 and 28 as assumed mean incase of x and y respectively:

x: 24 27 28 28 29 30 32 33 35 35 40
y: 18 20 22 25 22 28 28 30 27 30 32
Answers
1. 0.82 2. 0.603 4. 0.9804 5. 0.9158
6. 0.51 7. 0.84 8. 0.22, no 9. – 0.454
10. 0.8926.
www.cgaspirants.com
www.cgaspirants.com
3.38 REGRESSION ANALYSIS (U.P.T.U. 2015)
The term ‘regression’ was first used by Sir Francis Galton (1822–1911), a British Biometrician
in connection with the height of parents and their offsprings. He found that the offspring of
tall or short parents tend to regress to the average height. In other words, though tall fathers
do tend to have tall sons yet the average height of tall fathers is more than the average height
of their sons and the average height of short fathers is less than the average height of their
sons.
The term ‘regression’ stands for some sort of functional relationship between two or
more related variables. The only fundamental difference, if any, between problems of curve-
fitting and regression is that in regression, any of the variables may be considered as inde-
pendent or dependent while in curve-fitting, one variable cannot be dependent.
m
Regression measures the nature and extent of correlation. Regression is the estimation
or prediction of unknown values of one variable from known values of another variable.
o
3.39 CURVE OF REGRESSION AND REGRESSION EQUATION
.c
If two variates x and y are correlated i.e., there exists an association or relationship between
ts
them, then the scatter diagram will be more or less concentrated round a curve. This curve is
called the curve of regression and the relationship is said to be expressed by means of curvilinear
regression.
n
The mathematical equation of the regression curve is called regression equation.
ra
3.40 LINEAR REGRESSION

pi
When the points of the scatter diagram concentrate round a straight line, the regression is
called linear and this straight line is known as the line of regression.
as
Regression will be called non-linear if there exists a relationship other than a straight
line between the variables under consideration.
g
3.41 LINES OF REGRESSION (U.P.T.U. 2006, 2007)

.c
A line of regression is the straight line which gives the best fit in the least square sense to the
w
given frequency.
In case of n pairs (xi, yi) ; i = 1, 2, ..., n from a bivariate data, we have no reason or
w
justification to assume y as dependent variable and x as independent variable. Either of the

two may be estimated for the given values of the other. Thus if we wish to estimate y for given
w
values of x, we shall have the regression equation of the form y = a + bx, called the regression
line of y on x. If we wish to estimate x for given values of y, we shall have the regression line of
the form x = A + By, called the regression line of x on y.
Thus it implies, in general, we always have two lines of regression.
If the line of regression is so chosen that the sum of squares of deviation parallel to the
axis of y is minimised [See Fig. (a)], it is called the line of regression of y on x and it gives the
best estimate of y for any given value of x.
If the line of regression is so chosen that the sum of squares of deviations parallel to the
axis of x is minimised [See Fig. (b)], it is called the line of regression of x on y and it gives the
best estimate of x for any given value of y.
www.cgaspirants.com
www.cgaspirants.com
m
Fig. (a) Fig. (b)
The independent variable is called predictor or regresser or explanator and the dependent
variable is called the predictant or regressed or explained variable.
o
.c
3.42 DERIVATION OF LINES OF REGRESSION
3.42.1. Line of Regression of y on x

ts
To obtain the line of regression of y on x, we shall assume y as dependent variable and x
n
as independent variable. Let y = a + bx be the equation of regression line of y on x.
ra
The residual for ith point is Ei = yi – a – bxi.
pi
n n
U= ∑ Ei 2 = ∑ (y
i=1
i − a − bxi ) 2 ...(1)
i=1
as
According to the principle of Least squares, the constants a and b are chosen in such a
way that the sum of the squares of residuals is minimum.
g
Now, the condition for U to be maximum or minimum is

.c
∂U ∂U
=0 and =0
∂a ∂b
w
n
∂U
From (1),
∂a
=2 ∑ (yi – a – bxi)(– 1)
w
i=1
n
∂U
∑ (y – a – bx )(– 1) = 0
w
= 0 gives 2 i i
∂a i=1
⇒ Σy = na + b Σx ..(2)
n
∂U
Also,
∂b
=2
i=1
∑( yi − a − bxi )(− xi )
n
∂U
∂b
= 0 gives 2 ∑ (y – a – bx )(– x ) = 0
i=1
i i i
⇒ Σxy = a Σx + b Σx2 ...(3)
www.cgaspirants.com
www.cgaspirants.com
Equations (2) and (3) are called normal equations.

Solving (2) and (3) for ‘a’ and ‘b’, we get
1
Σxy − Σx Σy
n n Σxy − Σx Σy
b= = ...(4)
1 2 2
Σx 2 − (Σx) 2 n Σx − (Σx)
n
Σy Σx
and a= −b = y − bx ...(5)
n n
Eqn. (5) gives y = a + bx
Hence y = a + bx line passes through point ( x , y ) .
Putting a = y − bx in equation of line y = a + bx, we get
m
y – y = b( x − x ) ...(6)
o
Equation (6) is called regression line of y on x. ‘b’ is called the regression coefficient of y
.c
on x and is usually denoted by byx.
Hence eqn. (6) can be rewritten as
y – y = byx ( x − x )
where x and y are mean values while

n ts
ra
n Σxy − Σx Σy
byx =
n Σx 2 − (Σx) 2
pi
In equation (3), shifting the origin to ( x , y ) , we get

Σ(x – x )(y – y ) = a Σ(x – x ) + b Σ(x – x )2 ∵ Σ ( x − x ) = 0,
as
⇒ nr σxσy = a(0) + bnσx2 1

σy Σ( x − x ) 2 = σ x 2
⇒ b=r n
Σ ( x − x )( y − y)
g
σx and =r
Hence, regression coefficient byx can also be defined as nσ x σ y
.c
σy
w
byx = r
σx
w
where r is the coefficient of correlation, σx and σy are the standard deviations of x and y series
respectively.
w
3.42.2. Line of Regression of x on y

Proceeding in the same way as 3.42.1, we can derive the regression line of x on y as
x – x = bxy(y – y )
where bxy is the regression coefficient of x on y and is given by
n Σxy − Σx Σy
bxy =
n Σy 2 − (Σy) 2
σx
or bxy = r
σy
where the terms have their usual meanings.
www.cgaspirants.com
www.cgaspirants.com
Note. If r = 0, the two lines of regression become y = y and x = x which are two straight lines parallel to
x and y axes respectively and passing through their means y and x . They are mutually perpendicular.
If r = ± 1, the two lines of regression will coincide.
3.43 USE OF REGRESSION ANALYSIS (U.P.T.U. 2008)
(i) In the field of Business, this tool of statistical analysis is widely used. Businessmen are
interested in predicting future production, consumption, investment, prices, profits and sales
etc.
(ii) In the field of economic planning and sociological studies, projections of population,
birth rates, death rates and other similar variables are of great use.
m
3.44 COMPARISON OF CORRELATION AND REGRESSION ANALYSIS
[G.B.T.U. M.B.A. (C.O.) 2011]
o
Both the correlation and regression analysis helps us in studying the relationship between
.c
two variables yet they differ in their approach and objectives.
(i) Correlation studies are meant for studying the covariation of the two variables. They
ts
tell us whether the variables under study move in the same direction or in reverse directions.
The degree of their covariation is also reflected in the correlation co-efficient but the correla-
tion study does not provide the nature of relationship. It does not tell us about the relative
n
movement in the variables and we cannot predict the value of one variable corresponding to
ra
the value of other variable. This is possible through regression analysis.
(ii) Regression presumes one variable as a cause and the other as its effect. The
independent variable is supposed to be affecting the dependent variable and as such we can
pi
estimate the values of the dependent variable by projecting the relationship between them.
However, correlation between two series is not necessarily a cause-effect relationship.
as
(iii) Coefficient of correlation cannot exceed unity but one of the regression coefficients
can have a value higher than unity but the product of two regression coefficients can never
exceed unity.
g
.c
3.45 PROPERTIES OF REGRESSION CO-EFFICIENTS
Property I. Correlation co-efficient is the geometric mean between the regression co-efficients.
w
rσ y rσ x
Proof. The co-efficients of regression are and .
w
σx σy
rσ y rσ x
w
G.M. between them = × = r 2 = r = co-efficient of correlation.

σx σy
Property II. If one of the regression co-efficients is greater than unity, the other must be less
than unity.
rσ y rσ x
Proof. The two regression co-efficients are byx = and bxy = .
σx σy
1
Let byx > 1, then <1 …(1)
byx
Since byx. bxy = r2 ≤ 1 (∵ – 1 ≤ r ≤ 1)
1
∴ bxy ≤ < 1. | Using (1)
byx
Similarly, if bxy > 1, then byx < 1.
www.cgaspirants.com
www.cgaspirants.com
Property III. Arithmetic mean of regression co-efficients is greater than the correlation
co-efficient.
Proof. We have to prove that
byx + bxy
>r
2
σy σ
or r + r x > 2r
σx σy
or σx2 + σy2 > 2σxσy
or (σx – σy)2 > 0, which is true.
Property IV. Regression co-efficients are independent of the origin but not of scale.
m
x–a y−b
Proof. Let u= ,v= , where a, b, h and k are constants
h k
F I
o
rσ y kσ v k r σ v k
byx = =r. = GH JK
= bvu
.c
σx hσ u h σ u h
h
Similarly, bxy = b .
k uv
ts
Thus, byx and bxy are both independent of a and b but not of h and k.
n
Property V. The correlation co-efficient and the two regression co-efficients have same sign.
σ
ra
Proof. Regression co-efficient of y on x = byx = r y

σx
σ
pi
Regression co-efficient of x on y = bxy = r x

σy
Since σx and σy are both positive; byx, bxy and r have same sign.
as
3.46 ANGLE BETWEEN TWO LINES OF REGRESSION

g
If θ is the acute angle between the two regression lines in the case of two variables x and y, show
.c
that
w
1 − r2 σ xσ y
tan θ = . 2 , where r, σx, σy have their usual meanings.
r σx + σ y2
w
Explain the significance of the formula when r = 0 and r = ± 1.

[U.P.T.U. 2007, 2015; G.B.T.U. (C.O.) 2011]
w
Proof. Equations to the lines of regression of y on x and x on y are

rσ y rσ x
y− y= (x − x) and x − x = ( y − y)
σx σy
rσ y σy
Their slopes are m1 = and m2 = .
σx rσ x
σy rσ y
−
m2 − m1 rσ x σx
∴ tan θ = ± =±
1 + m2 m1 σ y2
1+
σ x2
www.cgaspirants.com
www.cgaspirants.com
1 − r2 σ y σ 2 1 − r2 σ xσ y
=± . . 2 x 2 =± . 2
r σx σx + σy r σ x + σ y2
Since r2 ≤ 1 and σx, σy are positive.
∴ +ve sign gives the acute angle between the lines.
1 − r2 σ xσ y
Hence tan θ = . 2
r σ x + σ y2
π
when r = 0, θ= ∴ The two lines of regression are perpendicular to each other.
2
Hence the estimated value of y is the same for all values of x and vice-versa.
m
When r = ± 1, tan θ = 0 so that θ = 0 or π
o
Hence the lines of regression coincide and there is perfect correlation between the two
variates x and y.
.c
EXAMPLES
coefficient of correlation?
n ts
Example 1. If the regression coefficients are 0.8 and 0.2, what would be the value of
ra
Sol. We know that,
r2 = byx . bxy = 0.8 × 0.2 = 0.16
pi
Since r has the same sign as both the regression coefficients byx and bxy
Hence r= 0.16 = 0.4.
as
Example 2. Calculate linear regression coefficients from the following:

g
x → 1 2 3 4 5 6 7 8
.c
y → 3 7 10 12 14 17 20 24
Sol. Linear regression coefficients are given by
w
n Σxy − Σx Σy n Σxy − Σx Σy
byx = and bxy =
w
2
n Σx − (Σx) 2
n Σy 2 − (Σy) 2
Let us prepare the following table:
w
x y x2 y2 xy
1 3 1 9 3
2 7 4 49 14
3 10 9 100 30
4 12 16 144 48
5 14 25 196 70
6 17 36 289 102
7 20 49 400 140
8 24 64 576 192
Σx = 36 Σy = 107 Σx2 = 204 Σy2 = 1763 Σxy = 599
www.cgaspirants.com
www.cgaspirants.com
Here, n = 8
(8 × 599) − (36 × 107) 940
∴ byx = 2 = = 2.7976
(8 × 204) − (36) 336
(8 × 599) − (36 × 107) 940
and bxy = 2
= = 0.3540
(8 × 1763) − (107) 2655
Example 3. The following table gives age (x) in years of cars and annual maintenance
cost (y) in hundred rupees:
X: 1 3 5 7 9
Y: 15 18 21 23 22
m
Estimate the maintenance cost for a 4 year old car after finding the regression equation.
Sol.
o
x y xy x2
.c
1 15 15 1
3
5
7
9
18
21
23
22
n ts
54
105
161
198
9
25
49
81
ra
Σx = 25 Σy = 99 Σxy = 533 Σx2 = 165

pi
Here, n=5
Σx 25 Σy 99
x= = = 5, y = = = 19.8
as
n 5 n 5
n Σxy − Σx Σy (5 × 533) − (25 × 99)
∴ byx = = = 0.95
g
n Σx 2 − (Σx) 2 (5 × 165) − (25) 2

Regression line of y on x is given by
.c
y − y = byx ( x − x )
w
⇒ y – 19.8 = 0.95 (x – 5)
w
⇒ y = 0.95x + 15.05
When x = 4 years, y = (0.95 × 4) + 15.05 = 18.85 hundred rupees = ` 1885.
w
Example 4. In a partially destroyed laboratory record of an analysis of a correlation

data, the following results only are legible:
Variance of x = 9
Regression equations: 8x – 10y + 66 = 0, 40x – 18y = 214.
What were (a) the mean values of x and y (b) the standard deviation of y and the
co-efficient of correlation between x and y? [U.P.T.U. 2008, 2009; U.K.T.U. 2010]
Sol. (a) Since both the lines of regression pass through the point ( x , y ) therefore,
we have
8 x − 10 y + 66 = 0 …(1)
40 x − 18 y − 214 = 0 …(2)
www.cgaspirants.com
www.cgaspirants.com
Multiplying (1) by 5, 40 x − 50 y + 330 = 0 …(3)
Subtracting (3) from (2), 32 y − 544 = 0 ∴ y = 17

∴ From (1), 8 x – 170 + 66 = 0 or 8 x = 104 ∴ x = 13
Hence, x = 13, y = 17
(b) Variance of x= σx2 =9 (given)
∴ σx = 3
The equations of lines of regression can be written as
y = 0.8x + 6.6 and x = 0.45y + 5.35
m
rσ y
∴ The regression co-efficient of y on x is = 0.8 …(4)
σx
o
rσ x
The regression co-efficient of x on y is = 0.45 …(5)
.c
σy
Multiplying (4) and (5), r2 = 0.8 × 0.45 = 0.36 ∴ r = 0.6
From (4), σy =
0.8σ x 0.8 × 3
= 4.
ts
(+ve sign with square root is taken because regression co-efficients are +ve).
n
=
ra
r 0.6
Example 5. The regression lines of y on x and x on y are respectively y = ax + b,
pi
x = cy + d. Show that
as
σy a bc + d ad + b
= , x= and y = .
σx c 1 − ac 1 − ac
g
[U.P.T.U. (C.O.) 2009, U.P.T.U. (MCA) 2008]

.c
Sol. The regression line of y on x is

y = ax + b …(1)
w
∴ byx = a
w
The regression line of x on y is

x = cy + d …(2)
w
∴ bxy = c
σy
We know that, byx = r …(3)
σx
σx
and bxy = r …(4)
σy
Dividing eqn. (3) by (4), we get
2
byx σ y2 a σy σy a
= ⇒ = 2 ⇒ =
bxy σx 2 c σx σx c
www.cgaspirants.com
www.cgaspirants.com
Since both the regression lines pass through the point ( x , y ) therefore,
y = ax + b and x = cy + d
⇒ ax − y = − b …(5)
x − cy = d …(6)
Multiplying equation (6) by a and then subtracting from (5), we get
ad + b
(ac – 1) y = − ad − b ⇒ y=
1 − ac
bc + d
Similarly, we get x= .
1 − ac
Example 6. For two random variables, x and y with the same mean, the two regression
m
b 1− a
equations are y = ax + b and x = αy + β. Show that = . Find also the common mean.
o
β 1− α
(G.B.T.U. 2010)
.c
Sol. Here, byx = a, bxy = α
Let the common mean be m, then regression lines are
⇒
y – m = a (x – m)
n
y = ax + m (1 – a)
ts …(1)
ra
and x – m = α(y – m)
⇒ x = αy + m (1 – α) …(2)
pi
Comparing (1) and (2) with the given equations.

b = m (1 – a), β = m (1 – α)
as
b 1− a
∴ =
β 1− α
g
Since regression lines pass through ( x , y )

.c
∴ x = αy + β and y = ax + b will hold.

⇒ m = am + b, m = αm + β
w
⇒ am + b = αm + β
w
β−b
⇒ m= .
a−α
w
Example 7. (i) Obtain the line of regression of y on x for the data given below:
x: 1.53 1.78 2.60 2.95 3.42
y: 33.50 36.30 40.00 45.80 53.50.
(ii) The following data regarding the heights (y) and weights (x) of 100 college students
are given:
Σx = 15000, Σx2 = 2272500, Σy = 6800, Σy2 = 463025 and Σxy = 1022250.
Find the equation of regression line of height on weight.
Sol. (i) The line of regression of y on x is given by
y – y = byx ( x − x ) ...(1)
www.cgaspirants.com
www.cgaspirants.com
where byx is the coefficient of regression given by

n Σxy − Σx Σy
byx = ...(2)
n Σx 2 − (Σx) 2
Now we form the table as,
x y x2 xy
1.53 33.50 2.3409 51.255
1.78 36.30 2.1684 64.614
2.60 40.00 6.76 104
2.95 45.80 8.7025 135.11
m
3.42 53.50 11.6964 182.97
Σx = 12.28 Σy = 209.1 Σx2 = 32.6682 Σxy = 537.949
o
Here, n=5
.c
(5 × 537.949) − (12.28 × 209.1)
byx = = 9.726
(5 × 32.6682) − (12.28) 2
Also, mean x =
Σx 12.28
n
=
5
n
= 2.456 and
ts
y=
Σy 2091
n
=
5
.
= 41.82
ra
∴ From (1), we get

y – 41.82 = 9.726(x – 2.456) = 9.726x – 23.887
pi
y = 17.932 + 9.726x
as
Σx 15000 Σy 6800
(ii) x= = = 150, y= = = 68
n 100 n 100
g
Regression coefficient of y on x,
.c
n Σxy − Σx Σy (100 × 1022250) − (15000 × 6800)

byx = = = 0.1
2
n Σx − (Σx) 2
(100 × 2272500) − (15000) 2
w
Regression line of height (y) on weight (x) is given by

w
y – y = byx ( x − x )
⇒ y – 68 = 0.1(x – 150)
w
⇒ y = 0.1x + 53.
Example 8. For 10 observations on price (x) and supply (y), the following data were
obtained (in appropriate units):
Σx = 130, Σy = 220, Σx2 = 2288, Σy2 = 5506 and Σxy = 3467
Obtain the two lines of regression and estimate the supply when the price is 16 units.
Σx Σy
Sol. Here, n = 10, x = = 13 and y= = 22
n n
Regression coefficient of y on x is
n Σxy − Σx Σy (10 × 3467) − (130 × 220)
byx = 2 2 = = 1.015
n Σx − (Σx) (10 × 2288) − (130) 2
www.cgaspirants.com
www.cgaspirants.com
∴ Regression line of y on x is
y – y = byx ( x − x )
y – 22 = 1.015(x – 13)
⇒ y = 1.015x + 8.805 ...(1)
Regression coefficient of x on y is
n Σxy − Σx Σy (10 × 3467) − (130 × 220)
bxy = = = 0.9114
2
n Σy − (Σy) 2
(10 × 5506) − (220) 2
Regression line of x on y is
x – x = bxy ( y − y)
m
x – 13 = 0.9114(y – 22)
x = 0.9114y – 7.0508 ...(2)
o
Since we are to estimate supply ( y) when price (x) is given therefore we are to use
.c
regression line of y on x here.
When x = 16 units,
ts
y = 1.015(16) + 8.805 = 25.045 units.
Example 9. The following results were obtained from records of age (x) and systolic
n
blood pressure (y) of a group of 10 men:
x y
ra
Mean 53 142
Variance 130
pi
and Σ(x – x )(y – y ) = 1220

Find the appropriate regression equation and use it to estimate the blood pressure of a
as
man whose age is 45.

Sol. Given
g
Mean x = 53 and y = 142; Variance σx2 = 130

.c
n = 10 and Σ(x – x )(y – y ) = 1220

w
Since we are to estimate blood pressure (y) of a 45 years old man, we will find regression
line of y on x.
w
Regression coefficient
F I
w
σy Σ ( x − x ) ( y − y) σ y
byx = r
σx
=
n σx σy
GH JK
σx
Σ ( x − x ) ( y − y) 1220
= 2
= = 0.93846
n σx (10) (130)
y − y = byx ( x − x )
⇒ y – 142 = 0.93846 (x – 53)
⇒ y = 0.93846 x + 92.26162
When x = 45, y = 134.49
Hence the required blood pressure = 134.49.
www.cgaspirants.com
www.cgaspirants.com
Example 10. The following results were obtained from marks in Applied Mechanics
and Engineering Mathematics in an examination:
Applied Mechanics (x) Engineering Mathematics (y)
Mean 47.5 39.5
Standard Deviation 16.8 10.8
r = 0.95.
Find both the regression equations. Also estimate the value of y for x = 30.
Sol. x = 47.5, y = 39.5
σx = 16.8, σy = 10.8 and r = 0.95.
Regression coefficients are
m
σy 10.8
byx = r = 0.95 × = 0.6107
σx 16.8
o
σx 16.8
= 0.95 ×
.c
and bxy = r = 1.477.
σy 10.8
Regression line of y on x is
⇒
y – y = byx ( x − x )
n ts
y – 39.5 = 0.6107 (x – 47.5) = 0.6107x – 29.008
ra
y = 0.6107x + 10.49 ...(1)
Regression line of x on y is
pi
x – x = bxy ( y − y)
⇒ x – 47.5 = 1.477 (y – 39.5)
as
x = 1.477y – 10.8415 ...(2)

Putting x = 30 in equation (1), we get
y = (0.6107)(30) + 10.49 = 28.81.
g
Example 11. The equations of two regression lines, obtained in a correlation analysis of
.c
60 observations are:
5x = 6y + 24 and 1000y = 768x – 3608.
w
What is the correlation coefficient? Show that the ratio of coefficient of variability of x to
w
5
that of y is . What is the ratio of variances of x and y?
24
w
Sol. Regression line of x on y is

5x = 6y + 24
6 24
⇒ x= y+
5 5
6
∴ bxy = ...(1)
5
Regression line of y on x is
1000y = 768x – 3608
⇒ y = 0.768x – 3.608
∴ byx = 0.768 ...(2)
www.cgaspirants.com
www.cgaspirants.com
σx 6
From (1), r = ...(3)
σy 5
σy
From (2), r = 0.768 ...(4)
σx
Multiplying equations (3) and (4), we get
r2 = 0.9216 ⇒ r = 0.96 ...(5)
Dividing (4) by (3), we get
σ x2 6
2
= = 1.5625.
σy 5 × 0.768
m
Taking square root, we get
σx 5
= 1.25 = ...(6)
o
σy 4
.c
Since the regression lines pass through the point ( x , y ), we have
5 x = 6 y + 24
1000 y = 768 x – 3608.
n
Solving the above equations for x and y , we get
ts x = 6 , y = 1.
ra
σx
Co-efficient of variability of x=
x
pi
σy
Co-efficient of variability of y= .
y
F I = 1×5 = 5 .
as
σx y y σx
∴ Required ratio =
x
×
σy
=
x σy GH JK 6 4 24 | Using (6)
g
Example 12. A panel of two judges, A and B, graded seven TV serial performances by
.c
awarding marks independently as shown in the following table:

w
Performance 1 2 3 4 5 6 7
w
Marks by A 46 42 44 40 43 41 45
Marks by B 40 38 36 35 39 37 41
w
The eighth TV performance which judge B could not attend, was awarded 37 marks by
judge A. If the judge B had also been present, how many marks would be expected to have been
awarded by him to the eighth TV performance?
Use regression analysis to answer this question.
Sol. Let the marks awarded by judge A be denoted by x and the marks awarded by
judge B be denoted by y.
Σx 46 + 42 + 44 + 40 + 43 + 41 + 45
Here, n = 7; x= = = 43
n 7
Σy 40 + 38 + 36 + 35 + 39 + 37 + 41
y= = = 38
n 7
www.cgaspirants.com
www.cgaspirants.com
Let us form the table as
x y xy x2
46 40 1840 2116
42 38 1596 1764
44 36 1584 1936
40 35 1400 1600
43 39 1677 1849
41 37 1517 1681
45 41 1845 2025
m
Σx = 301 Σy = 266 Σxy = 11459 Σx2 = 12971
o
Regression coefficient,
.c
n Σxy − Σx Σy (7 × 11459) − (301 × 266)
byx = 2 2 = = 0.75
n Σx − (Σx) (7 × 12971) − (301) 2
n
y – y = byx ( x − x )
ts
y – 38 = 0.75(x – 43)
ra
⇒ y = 0.75x + 5.75
when x = 37, y = 0.75(37) + 5.75 = 33.5 marks
pi
Hence, if judge B had also been present, 33.5 marks would be expected to have been
awarded to the eighth TV performance.
as
Example 13. Two variables x and y have zero means, the same variance σ2 and zero
correlation, show that:
g
u = x cos α + y sin α and v = x sin α – y cos α

have the same variance σ2 and zero correlation.
.c
(U.P.T.U. 2007)
Sol. We are given that
w
r (x, y) = 0 ⇒ Cov (x, y) = 0, σx2 = σy2 = σ2

We have, σu2 = V (x cos α + y sin α)
w
= cos2 α V (x) + sin2 α V(y) + 2 sin α cos α Cov (x, y)

= (cos2 α + sin2 α) σ2 |∵ Cov (x, y) = 0
w
= σ2
Similarly, σv2 = σ2
Cov (u, v) = E [(u – u ) (v – v )]
= E [(x cos α + y sin α – x cos α – y sin α)
(x sin α – y cos α – x sin α + y cos α)]
= E[(x cos α + y sin α) (x sin α – y cos α)] |∵ x =0= y
= [E(x2)
– E(y2)]
sin α cos α + E(xy) α– (sin2
α) cos2
=0 2 2 2
|∵ σx = σy = σ and E(xy) = 0
Cov (u, v)
∴ r= = 0.
σu σv
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. (i) Discuss regression and its importance. Given the following data:
x: 1 5 3 2 1 1 7 3
y: 6 1 0 0 1 2 1 5
Find a regression line of x on y. (U.P.T.U. 2008)
(ii) In a study between the amount of rainfall and the quantity of air pollution removed the
following data were collected:
Daily rainfall: 4.3 4.5 5.9 5.6 6.1 5.2 3.8 2.1
(in .01 cm)
Pollution removed: 12.6 12.1 11.6 11.8 11.4 11.8 13.2 14.1
m
(mg/m3)
Find the regression line of y on x.
o
(iii) Find the two lines of regression and coefficient of correlation for the data given below:
n = 18, Σ x = 12, Σ y = 18, Σ x2 = 60, Σ y2 = 96, Σ xy = 48 [U.P.T.U. (MCA) 2009]
.c
(iv) From the data given, find the equation of lines of regression of x on y and y on x. Also calculate
the correlation co-efficient.
2.
y:
x:
5
2 4
7
6
n
9
8 10
8 ts 11
(i) Can Y = 5 + 2.8 X and X = 3 – 0.5 Y be the estimated regression equations of Y on X and X on
(U.P.T.U. 2011)
ra
Y respectively? Explain your answer with suitable theoretical arguments.

(ii) Find the co-efficient of correlation when the two regression equations are
X = – 0.2 Y + 4.2, Y = – 0.8 X + 8.4
pi
3. (i) If F is the pull required to lift a load W by means of a pulley block, fit a linear law of the
form F = mW + c connecting F and W, using the data
as
W: 50 70 100 120
F: 12 15 21 25
g
where F and W are in kg wt. Compute F when W = 150 kg wt. (U.P.T.U. 2007)
.c
(ii) A simply supported beam carries a concentrated load P (kg) at its mid-point. The following
table gives maximum deflection y (cm) corresponding to various values of P:
w
P: 100 120 140 160 180 200

y: 0.45 0.55 0.60 0.70 0.80 0.85
w
Find a law of the form y = a + bP. Also find the value of maximum deflection when P = 150 kg.
4. (i) Find both the lines of regression of following data:
w
x: 5.60 5.65 5.70 5.81 5.85

y: 5.80 5.70 5.80 5.79 6.01
(ii) Obtain regression line of x on y for the given data:
x: 1 2 3 4 5 6
y: 5.0 8.1 10.6 13.1 16.2 20.0 [U.P.T.U. (MCA) 2007]
(iii) Given that:
x: 1 3 5 7 8 10
y: 8 12 15 17 18 20
Find the equations of both lines of regression. [U.P.T.U. (C.O.) 2008]
www.cgaspirants.com
www.cgaspirants.com
5. (i) The two regression equations of the variables x and y are x = 19.13 – 0.87 y and y = 11.64 – 0.50x.
Find (a) mean of x’s (b) mean of y’s and (c) correlation coefficient between x and y.
(ii) Two random variables have the regression lines with equations 3x + 2y = 26 and 6x + y = 31.
Find the mean values and the correlation coefficient between x and y.
[G.B.T.U. (MBA) 2011]
(iii) In a partially destroyed laboratory data, only the equations giving the two lines of regression
of y on x and x on y are available and are respectively
7x – 16y + 9 = 0, 5y – 4x – 3 = 0.
Calculate the coefficient of correlation, x and y .
(iv) The regression equations calculated from a given set of observations for two random vari-
ables are
m
x = – 0.4y + 6.4 and y = – 0.6x + 4.6
Calculate (i) x (ii) y (iii) r.
o
(v) Two lines of regression are given by
x + 2y – 5 = 0 and 2x + 3y – 8 = 0 and σx2 = 12,
.c
Calculate:
ts
(a) the mean values of x and y (b) variance of y
(c) the coefficient of correlation between x and y. [U.P.T.U. (MCA) 2008, G.B.T.U. (C.O.) 2011]
n
6. An analyst for a company was studying travelling expenses (y) in ` and duration (x) of these trips
for 102 sales trip. He has found relation between x and y linear and data as follows:
ra
Σx = 510, Σy = 7140, Σx2 = 4150, Σxy = 54900, Σy2 = 740200

Calculate (i) Two regression lines
pi
(ii) A given trip has to take 7 days. How much money should be allowed so that they
will not run short of money?
as
7. Assuming that we conduct an experiment with 8 fields planted with corn, four fields having no
nitrogen fertiliser and four fields having 80 kgs of nitrogen fertilizer. The resulting corn yields
are shown in table in bushels per acre :
g
Field: 1 2 3 4 5 6 7 8
.c
Nitrogen (kgs) x: 0 0 0 0 80 80 80 80
Corn yield y: 120 360 60 180 1280 1120 1120 760
w
(acre)
(a) Compute a linear regression equation of y on x.
w
(b) Predict corn yield for a field treated with 60 kgs of fertilizer.
w
8. If the coefficient of correlation between two variables x and y is 0.5 and the acute angle between
their lines of regression is tan −1

FG 3 IJ , show that σ =
1
σy. (U.P.T.U. 2009)
H 5K x
2
9
9. Given N = 50, Mean of y = 44, Variance of x is of the variance of y.
16
Regression equation of x on y is 3y – 5x = – 180
Find (i) Mean of x (ii) Coeff. of correlation between x and y.
10. The means of a bivariate frequency distribution are at (3, 4) and r = 0.4. The line of regression of
y on x is parallel to the line y = x. Find the two lines of regression and estimate value of x when
y = 1.
www.cgaspirants.com
www.cgaspirants.com
11. The following results were obtained in the analysis of data on yield of dry bark in ounces (y) and
age in years (x) of 200 cinchona plants:
x y
Average: 9.2 16.5
Standard deviation: 2.1 4.2
Correlation coefficient = 0.84
Construct the two lines of regression and estimate the yield of dry bark of a plant of age 8 years.
12. A panel of judges A and B graded 7 debators and independently awarded the following marks:
Debator: 1 2 3 4 5 6 7
Marks by A: 40 34 28 30 44 38 31
m
Marks by B: 32 39 26 30 38 34 28
An eighth debator was awarded 36 marks by judge A while judge B was not present. If judge B
o
were also present, how many marks would you expect him to award to the eighth debator assum-
ing that the same degree of relationship exists in their judgement?
.c
Answers
1. (i) 72 x = – 20 y + 247
ts
(ii) y = – 0.6842x + 15.5324
(iii) y = 0.6923 x + 0.53846 ; x = 0.4615 y + 0.2051
n
(iv) x = 1.3 y – 4.4, y = .65x + 4.1; r = .9192
ra
2. (i) No (ii) r = – 0.4
3. (i) F = 0.18793W + 2.27595; F = 30.4654 kg wt.
pi
(ii) y = 0.04765 + 0.004071 P; y = 0.6583 cm

4. (i)Regression line of y on x: y = 0.74306 x + 1.56821
as
Regression line of x on y: x = 0.63602 y + 2.0204

(ii) x = 0.34195 y – 0.660355 (iii) y = 1.3012 x + 7.6265 ; x = 0.75y – 5.5833.
g
5. (i) (a) 15.935 (b) 3.67 (c) – 0.659

.c
(ii) x = 4, y = 7, r = – 0.5 (iii) r = 0.7395, x = – 0.1034, y = 0.5172

(iv) x = 6, y = 1, r = – 0.48989
w
3
(v) (a) x = 1, y = 2 (b) 4 (c) –
2
w
6. (i) y = 12x + 10, x = 0.07986y – 0.59068 (ii) ` 94

w
7. (a) y = 11.125x + 180 (b) 847.5 acre

9. (i) 62.4 (ii) 0.8
10. y = x + 1 ; x = 0.16y + 2.36 ; x = 2.52
11. y = 1.68x + 1.044, x = 0.42y + 2.27 ; y = 14.484
12. 33 marks.
3.47 POLYNOMIAL FIT: NON-LINEAR REGRESSION
Let y = a + bx + cx2 ...(1)

be a second degree parabolic curve of regression of y on x to be fitted for the data (xi, yi),
i = 1, 2, ......, n.
www.cgaspirants.com
www.cgaspirants.com
Residual at x = xi is
Ei = yi – f(xi) = yi – a – bxi – cxi2
n n
Now, let U= ∑
i= 1
Ei 2 = ∑
i= 1
(yi – a – bxi – cxi2)2
By principle of Least squares, U should be minimum for the best values of a, b and c.
∂U ∂U ∂U
For this, = 0, = 0 and =0
∂a ∂b ∂c
n
∂U
=0 ⇒ 2 ∑ (yi – a – bxi – cxi2) (– 1) = 0
m
∂a i= 1
o
⇒ Σy = na + bΣx + cΣx2 ...(1)
.c
n
∂U
=0 ⇒ 2 ∑ (yi – a – bxi – cxi2) (– xi) = 0
∂b n
⇒
i= 1
ts
Σxy = aΣx + bΣx2 + cΣx3 ...(2)
ra
n
∂U
=0 ⇒ 2 ∑ (yi – a – bxi – cxi2) (– xi2) = 0
pi
∂c i= 1
as
⇒ Σx2y = aΣx2 + bΣx3 + cΣx4 ...(3)
Equations (1), (2) and (3) are the normal equations for fitting a second degree parabolic
g
curve of regression of y on x. Here n is the no. of pairs of values of x and y.

.c
w
EXAMPLES
w
Example 1. (a) Fit a second degree parabola to the following data:

w
x 0.0 1.0 2.0

y 1.0 6.0 17.0
(b) Fit a second degree curve of regression of y on x to the following data:
x 1.0 2.0 3.0 4.0

y 6.0 11.0 18.0 27
(c) Fit a second degree parabola in the following data:
x 0.0 1.0 2.0 3.0 4.0

y 1.0 4.0 10.0 17.0 30.0
www.cgaspirants.com
www.cgaspirants.com
Sol. The equation of second degree parabola is given by

y = a + bx + cx2 ...(1)
Σy = ma + bΣx + cΣx2 ...(2)
Σxy = aΣx + bΣx2 + cΣx3 ...(3)
and Σx2y = aΣx2 + bΣx3 + cΣx4 ...(4)
(a) Here, m = 3. Table is as follows:
0 1 0 0 0 0 0
m
1 6 1 1 1 6 6
2 17 4 8 16 34 68
o
Total 3 24 5 9 17 40 74
.c
Substituting in eqns. (2), (3) and (4), we get
24 = 3a + 3b + 5c ...(5)
40 = 3a + 5b + 9c
74 = 5a + 9b + 17c
n
Solving eqns. (5), (6) and (7), we get a = 1, b = 2, c = 3
ts ...(6)
...(7)
ra
Hence the required second degree parabola is y = 1 + 2x + 3x2
(b) Here, m = 4. Table is as follows:
pi
1 6 1 1 1 6 6
as
2 11 4 8 16 22 44
3 18 9 27 81 54 162
4 27 16 64 256 108 432
g
Σx = 10 Σy = 62 Σx2 = 30 Σx3 = 100 Σx4 = 354 Σxy = 190 Σx2y = 644

.c
Substituting values in eqns. (2), (3) and (4), we get

w
62 = 4a + 10b + 30c ...(8)

190 = 10a + 30b + 100c ...(9)
w
644 = 30a + 100b + 354c ...(10)

Solving equations (8), (9) and (10), we get a = 3, b = 2, c = 1
w
Hence the required second degree parabola is y = 3 + 2x + x2

(c) Here, m = 5. Table is as follows:
0.0 1.0 0 0 0 0 0
1.0 4.0 1 1 1 4 4
2.0 10.0 4 8 16 20 40
3.0 17.0 9 27 81 51 153
4.0 30.0 16 64 256 120 480
www.cgaspirants.com
www.cgaspirants.com
Substituting values in eqns. (2), (3) and (4), we get

62 = 5a + 10b + 30c ...(11)
195 = 10a + 30b + 100c ...(12)
677 = 30a + 100b + 354c ...(13)
Solving eqns. (11), (12) and (13), we get a = 1.2, b = 1.1 and c = 1.5
Hence the required second degree parabola is y = 1.2 + 1.1x + 1.5x2.
Example 2. Fit a second degree parabola to the following data taking y as dependent
variable:
x 1 2 3 4 5 6 7 8 9
m
y 2 6 7 8 10 11 11 10 9
Sol. Normal equations to fit a second degree parabola of the form y = a + bx + cx2 are
o
Σy = ma + bΣx + cΣx 2 U| …(1)
.c
and Σxy = aΣx + bΣx 2 + cΣx 3 V|
Σx 2 y = aΣx 2 + bΣx 3 + cΣx 4 W
Here, m = 9
x y x2
n
x3
tsx4 xy x 2y
ra
1 2 1 1 1 2 2
2 6 4 8 16 12 24
3 7 9 27 81 21 63
pi
4 8 16 64 256 32 128
5 10 25 125 625 50 250
6 11
as
36 216 1296 66 396

7 11 49 343 2401 77 539
8 10 64 512 4096 80 640
g
9 9 81 729 6561 81 729

.c
w
Putting in (1), we get

74 = 9a + 45b + 285c
w
421 = 45a + 285b + 2025c

2771 = 285a + 2025b + 15333c
w
Solving the above equations, we get a = – 1, b = 3.55, c = – 0.27

Hence the required equation of second degree parabola is y = – 1 + 3.55x – 0.27x2.
Example 3. Employ the method of least squares to fit a parabola y = a + bx + cx2 in the
data: (x, y): (– 1, 2), (0, 0), (0, 1), (1, 2)
Sol. Normal equations to the parabola y = a + bx + cx2 are
Σy = ma + bΣx + cΣx2 ...(1)
2
Σxy = aΣx + bΣx + cΣx 3 ...(2)
and 2 2 3
Σx y = aΣx + bΣx + cΣx 4 ...(3)
www.cgaspirants.com
www.cgaspirants.com
Here, m = 4. The table is as follows:
–1 2 1 –1 1 –2 2
0 0 0 0 0 0 0
0 1 0 0 0 0 0
1 2 1 1 1 2 2
Substituting these values in equations (1), (2) and (3); we get

5 = 4a + 2c ...(4)
m
0 = 2b ...(5)
and 4 = 2a + 2c ...(6)
Solving (4), (5) and (6), we get a = 0.5, b = 0 and c = 1.5
o
Hence the required second degree parabola is y = 0.5 + 1.5x2
.c
Example 4. Fit a second degree parabola to the following data by Least Squares method:
ts
x 1 2 3 4 5
y 1090 1220 1390 1625 1915
n
[U.P.T.U. (MCA) 2009, U.P.T.U. 2007; U.K.T.U. 2010]
ra
Sol. Here m = 5 (odd)

Let u = x – 3, v = y – 1220
pi
x y u v u2 u2 v uv u3 u4
as
1 1090 –2 – 130 4 – 520 260 –8 16

2 1220 –1 0 1 0 0 –1 1
3 1390 0 170 0 0 0 0 0
g
4 1625 1 405 1 405 405 1 1

.c
5 1915 2 695 4 2780 1390 8 16

w
Total Σu = 0 Σv = 1140 Σu2 = 10 Σu2v = 2665 Σuv = 2055 Σu3 = 0 Σu4 = 34
Putting these values in normal equations, we get

w
1140 = 5a′ + 10c′, 2055 = 10b′, 2655 = 10a′ + 34c′

w
⇒ a′ = 173, b′ = 205.5, c′ = 27.5

∴ v = 173 + 205.5u + 27.5u2 …(1)
Put u=x–3 and v = y – 1220
From (1), y – 1220 = 173 + 205.5 (x – 3) + 27.5 (x – 3)2
⇒ y = 27.5x2 + 40.5x + 1024.
3.48 MULTIPLE LINEAR REGRESSION
Now we proceed to discuss the case where the dependent variable is a function of two or more
linear or non-linear independent variables. Consider such a linear function as
y = a + bx + cz …(1)
www.cgaspirants.com
www.cgaspirants.com
The sum of the squares of residual is

n
U= ∑ (y
i=1
i − a − bxi − czi ) 2 …(2)
Differentiating U partially w.r.t. a, b, c ; we get

n
∂U
∂a
=0 ⇒ 2 ∑ (y
i=1
i − a − bxi − czi ) (– 1) = 0
n
∂U
∂b
=0 ⇒ 2 ∑ (y
i=1
i − a − bxi − czi ) (– xi) = 0
m
n
∂U
and
∂c
=0 ⇒ 2 ∑ (y
i=1
i − a − bxi − czi ) (– zi) = 0
o
which on simplification and omitting the suffix i, yields.
.c
∑y = ma + b∑x + c∑z
∑xy = a∑x + b∑x2 + c∑xz
∑yz = a∑z + b∑xz + c∑z2
ts
Solving the above three equations, we get values of a, b and c. Consequently, we get the
n
linear function y = a + bx + cz called regression plane.
ra
EXAMPLES
pi
Example 1. Obtain a regression plane by using multiple linear regression to fit the data
as
given below:
x 1 2 3 4
g
z 0 1 2 3
.c
y 12 18 24 30 [U.P.T.U. MCA (C.O.) 2008]

w
Sol. Let y = a + bx + cz be the required regression plane where a, b, c are the constants
to be determined by following equations:
w
Σy = ma + bΣx + cΣz
w
Σyx = aΣx + bΣx2 + cΣzx

and Σyz = aΣz + bΣzx + cΣz2
Here, m=4
x z y x2 z2 yx zx yz
1 0 12 1 0 12 0 0
2 1 18 4 1 36 2 18
3 2 24 9 4 72 6 48
4 3 30 16 9 120 12 90
Σx = 10 Σz = 6 Σy = 84 Σx2 = 30 Σz2 = 14 Σyx = 240 Σzx = 20 Σyz = 156
www.cgaspirants.com
www.cgaspirants.com
Substitution yields, 84 = 4a + 10b + 6c

240 = 10a + 30b + 20c
and 156 = 6a + 20b + 14c
Solving, we get a = 10, b = 2, c = 4
Hence the required regression plane is y = 10 + 2x + 4z
Example 2. Find the multiple linear regression of X1 on X2 and X3 from the data relating
to three variables:
X1 4 6 7 9 13 15
X2 15 12 8 6 4 3
m
X3 30 24 20 14 10 4
Sol. Let X1 = a + bX2 + cX3 be the required regression plane where a, b, c are the
o
constants, determined by following normal equations
.c
ΣX1 = ma + bΣX2 + cΣX3
ΣX1X2 = aΣX2 + bΣX22 + cΣX2X3
Here, m = 6
ΣX1X3 = aΣX3 + bΣX2X3 + cΣX32
n ts
ra
X1 X2 X3 X 1X 2 X 22 X 2X 3 X 1X 3 X 32
4 15 30 60 225 450 120 900

pi
6 12 24 72 144 288 144 576

7 8 20 56 64 160 140 400
as
9 6 14 54 36 84 126 196
13 4 10 52 16 40 130 100
g
15 3 4 45 9 12 60 16
.c
Total 54 48 102 339 494 1034 720 2188
Substituting the values, we get

w
54 = 6a + 48b + 102c
w
339 = 48a + 102b + 1034c

720 = 102a + 1034b + 2188c
w
On solving, we get, a = 16.413, b = – 0.00536, c = – 0.4335

Hence X1 = 16.413 – 0.00536X2 – 0.4335X3
ASSIGNMENT
1. Fit a parabola of the form y = a + bx + cx2 to the data:
x 1 2 3 4
y 1.7 1.8 2.3 3.2
by the method of least squares. (U.P.T.U. 2009)
www.cgaspirants.com
www.cgaspirants.com
2. Find the best values of a0, a1, a2 so that the parabola y = a0 + a1x + a2x2 fits the data:
x 1 1.5 2 2.5 3 3.5 4

y 1.1 1.2 1.5 2.6 2.8 3.3 4.1
[U.P.T.U. (C.O.) 2008]
3. (i) Fit a second degree parabola to the following data:
x 1 2 3 4 5
y 25 28 33 39 46
[U.P.T.U. (C.O.) 2011]
(ii) Fit a second degree parabola to the following data:
m
x 1 2 3 4 5 6 7 8 9 10
o
y 124 129 140 159 228 289 315 302 263 210
.c
(U.P.T.U. 2009)
4. Fit a second degree parabola to the following data taking x as the independent variable:
(i) x
y
0
1
n 1
5
ts 2
10
3
22
4
38
ra
(ii) x 1 2 3 4 5 6 7 8 9
pi
y 3 7 8 9 11 12 13 14 15
(U.P.T.U. 2007)
as
5. The profit of a certain company in Xth year of its life are given by:
x 1 2 3 4 5
g
y 1250 1400 1650 1950 2300

.c
y − 1650
Taking u = x – 3 and v =
, show that the parabola of second degree of v on u is
w
50
2
v + 0.086 = 5.3 u + 0.643u and deduce that the parabola of second degree of y on x is
w
y = 1144 + 72x + 32.15x2

6. (i) The corresponding values of x and y are given below:
w
x 87 84 79 64 47 37
y 292 283 270 235 197 181
Fit a parabola of the form y = ax2 + bx + c. Also find the value of y for x = 80 correct upto third
place of decimal. (U.P.T.U. 2006)
(ii) Determine the constants a, b and c by the method of least squares such that y = ax2 + bx + c
fits the following data:
x 2 4 6 8 10
y 4.01 11.08 30.12 81.89 222.62
www.cgaspirants.com
www.cgaspirants.com
7. The velocity V of a liquid is known to vary with temperature T, according to a quadratic law V = a
+ bT + cT2. Find the best values of a, b and c for the following table:
T 1 2 3 4 5 6 7
V 2.31 2.01 1.80 1.66 1.55 1.47 1.41

[G.B.T.U. (MCA) 2010]
8. The following table gives the results of the measurements of train resistances, V is the velocity in
miles per hour, R is the resistance in pounds per ton:
V 20 40 60 80 100 120
R 5.5 9.1 14.9 22.8 33.3 46
m
If R is related to V by the relation R = a + bV + cV2; find a, b and c by using the method of least
squares.
o
9. Find the multiple linear regression of X1 on X2 and X3 from the data relating to three variables:
.c
X1 7 12 17 20
X2
X3
4
1
n 7
2 ts
9
5
12
8 (U.P.T.U. 2009)
10. Fit a second degree parabola to the following data:

ra
x 1 2 3 4 5 6 7 8 9
pi
y 2 6 7 8 10 11 8 13 5
(U.P.T.U. 2015)
as
Answers
g
1. y = 2 – 0.5 x + 0.2 x2 2. y = 0.45714 + 0.39286 x + 0.12857 x2

.c
3. (i) y = 22.8 + 1.44x + 0.64x2 (ii) y = 18.866 + 66.1576 x – 4.3333 x2

4. (i) y = 1.43 + 0.24x + 2.21x2 (ii) y = 1.5238 + 2.38398 x – 0.10173 x2
w
6. (i) y = 0.010626822 x2 + 0.908257322 x + 132.2040143 ; 272.876

(ii) a = 5.358035714, b = – 38.89492857, c = 67.56
w
7. V = 2.5928 – 0.3258 T + 0.02274 T2 8. R = 4.35 + 0.00241 V + 0.0028705 V2

w
9. X1 = 0.6441 + 1.661X2 + 0.0169 X3 10. y = – 1.619 + 4.031 x – 0.339 x2.
3.49 THEORETICAL PROBABILITY DISTRIBUTIONS
Generally, frequency distribution are formed from the observed or experimental data. However,
frequency distribution of certain populations can be deduced mathematically by fitting
theoretical probability distribution under certain assumptions.
Frequency distributions can be classified under two heads:
(i) Observed Frequency Distributions.
(ii) Theoretical or Expected Frequency Distributions.
Observed frequency distributions are based on actual observation and experimentation.
If certain hypothesis is assumed, it is sometimes possible to derive mathematically what
www.cgaspirants.com
www.cgaspirants.com
the frequency distribution of certain universe should be. Such distributions are called
Theoretical Distributions.
Theoretical probability distributions are of two types:
(i) Discrete probability distribution. Binomial, poisson, geometric, negative binomial,
hypergeometric, multinomial, multivariate hypergeometric distributions.
(ii) Continuous probability distributions.
Uniform, normal Gamma, exponential, χ2, Beta, bivariate normal, t, F-distributions.
Here, we will study three important theoretical probability distributions:
1. Binomial Distribution (or Bernoulli’s Distribution)
2. Poisson’s Distribution
m
3. Normal Distribution.
o
3.50 BINOMIAL PROBABILITY DISTRIBUTION [G.B.T.U. 2010, 2013]
.c
It was discovered by a Swiss Mathematician Jacob James Bernoulli in the year 1700.
This distribution is concerned with trials of a repretitive nature in which only the occur-
event is of interest.
n ts
rence or non-occurrence, success or failure, acceptance or rejection, yes or no of a particular
For convenience, we shall call the occurrence of the event ‘a success’ and its non-occur-
ra
rence ‘a failure’.
Let there be n independent trials in an experiment. Let a random variable X denote the
number of successes in these n trials. Let p be the probability of a success and q that of a
pi
failure in a single trial so that p + q = 1. Let the trials be independent and p be constant for
every trial.
as
Let us find the probability of r successes in n trials.

r successes can be obtained in n trials in nCr ways.
g
∴ P(X = r) = n C r P(S S S ...... S F F F ...... F

.c
r times ( n – r ) times
= n C r P(S) P(S) ...... P(S) P(F) P(F) ...... P(F)

w
r factors ( n – r ) factors
w
n
= C r p p p ...... p q q q ...... q
r factors ( n – r ) factors
w
= nC prqn–r ...(1)
r
Hence P(X = r) = nCr pr qn–r where p + q = 1 and r = 0, 1, 2, ..., n.
The distribution (1) is called the binomial probability distribution and X is called the
binomial variate.
Note 1. P(X = r) is usually written as P(r).
Note 2. The successive probabilities P(r) in (1) for r = 0, 1, 2, ......, n are
nC n, nC n–1 p, nC2qn–2p2, ......, nCnpn
0q 1q
which are the successive terms of the binomial expansion of (q + p)n. That is why this distribution is
called “binomial” distribution.
www.cgaspirants.com
www.cgaspirants.com
Note 3. n and p occurring in the binomial distribution are called the parameters of the distribution.
Note 4. In a binomial distribution:
(i) n, the number of trials is finite.
(ii) each trial has only two possible outcomes usually called success and failure.
(iii) all the trials are independent.
(iv) p (and hence q) is constant for all the trials.
3.51 RECURRENCE OR RECURSION FORMULA FOR THE BINOMIAL DISTRIBUTION
In a binomial distribution,
n!
P(r) = nCr qn–r pr = qn–r pr
m
(n − r ) ! r !
n!
P(r + 1) = nCr+1 qn–r–1 pr+1 = qn–r–1 pr+1
o
(n − r − 1) ! (r + 1) !
P(r + 1) (n − r ) ! r! p
.c
∴ = × ×
P ( r) (n − r − 1) ! (r + 1) ! q
FG IJ
=
(n − r) × (n − r − 1) !
(n − r − 1) !
n
n−r p
×
r!
ts×
(r + 1) × r ! q
p
×=
n−r p
.
r+1 q H K
⇒ P(r + 1) = . P(r)
ra
r+1 q
which is the required recurrence formula. Applying this formula successively, we can find
pi
P(1), P(2), P(3), ......, if P(0) is known.
3.52 MEAN AND VARIANCE OF THE BINOMIAL DISTRIBUTION

as
[U.P.T.U. 2008, G.B.T.U. 2012]

g
For the binomial distribution, P(r) = nCr qn–r pr

.c
n n
Mean μ = ∑ rP(r) = ∑r. nC
rq
n–r pr
w
r=0 r=0
= 0 + 1 . nC1 qn–1 p + 2 . nC2 qn–2 p2 + 3 . nC3 qn–3 p3 + ...... + n . nCn pn
w
n(n − 1) n–2 2 n(n − 1)(n − 2) n–3 3

= nqn–1 p + 2 . q p +3. q p + ...... + n.pn
2.1 3 . 2 .1
w
n(n − 1)(n − 2) n–3 3

= nqn–1p + n(n – 1) qn–2p2 + q p + ...... + npn
2.1
= np[n–1C0qn–1 + n–1C1qn–2p + n–1C2qn–3p2 + ... + n–1Cn–1 pn–1]
= np(q + p)n–1 = np (∵ p + q = 1)
Hence the mean of the binomial distribution in np.
n n
Variance σ2 = ∑
r=0
r 2 P(r) − μ 2 = ∑ [r + r(r – 1)] P(r) – μ
r=0
2
n n n
= ∑ rP(r) + ∑ r(r – 1) P(r) – μ
r=0 r=0
2 =μ+ ∑ r(r – 1)
r=2
nC
rq
n–rpr – μ2
(since the contribution due to r = 0 and r = 1 is zero)
www.cgaspirants.com
www.cgaspirants.com
= μ + [2 .1 . nC2qn–2p2 + 3.2 . nC3qn–3p3 + ... + n(n – 1) nCnpn] – μ2

LM
= μ + 2 . 1.
n(n − 1) n− 2 2
q p + 3.2.
n(n − 1)(n − 2) n− 3 3
q p + ... + n(n − 1) pn – μ2
OP
N 2.1 3.2.1 Q
= μ + [n(n – 1)qn–2p2 + n(n – 1)(n – 2)qn–3p3 + ... + n(n – 1)pn] – μ2
= μ + n(n – 1) p2[qn–2 + (n – 2)qn–3p + ... + pn–2] – μ2
= μ + n (n – 1)p2 [n–2C0qn–2 + n–2C1qn–3p + ... + n–2Cn–2pn–2] – μ2
= μ + n(n – 1)p2(q + p)n–2 – μ2 = μ + n(n – 1)p2 – μ2 [∵ q + p = 1]
= np + n(n – 1)p2 – n2p2 = np[1 – p] = npq. [∵ μ = np]
Hence the variance of the binomial distribution is npq.
m
Standard deviation of the binomial distribution is npq .
3.53 MOMENT GENERATING FUNCTION OF BINOMIAL DISTRIBUTION
o
.c
1. About origin
n n
∑ ∑
ts
e tx n C x p x q n − x = n
Mx(t) = E(etx) = C x ( pe t ) x q n− x = (q + pe t ) n
x=0 x=0
n
2. About mean [G.B.T.U. 2012, U.P.T.U. 2008, 2015]
Mx–np (t) = E[et(x – np)]
ra
= e–npt E(etx) = e–npt Mx(t) = e–npt (q + pet)n

= (qe–pt + pet–pt)n = (qe–pt + peqt)n |∵ 1 – p = q
pi
3.54 MOMENTS ABOUT MEAN OF BINOMIAL DISTRIBUTION

as
Mx–np(t) = (qe–pt + peqt)n

LM F I F n
I OP
g
2 2 3 3 2 2 3 3
= q 1 − pt + p t − p t + ... + p 1 + qt + q t + q t + ...
MN GH JK GH JK P
Q
.c
2! 3! 2! 3!
L 2 3 4 O n
w
= M(q + p) + t pq (q + p) + t pq (q − p ) + t pq (q + p ) + ...P
2 2 3 3
MN 2! 3! 4! PQ
w
L R
= M1 + |S t . pq + t pq (q − p) + t qp (1 − 3 pq) + ...|VP
2 3 4 UO n
w
MN T| 2 ! 3! 4! W|PQ
L |R t
= M1 + C S . pq +
n
2
t 3
pq (q − p) +
t 4
|U
pq (1 − 3 pq) + ...V
MN 1
|T 2 ! 3! 4! |W
R| t t 2 U|
3 2 OP
+ C S . pq + n
pq (q − p) + ...V
T| 2 ! 2
3! W|
+ ...
PPQ
Now,
t2
μ2 = coefficient of = npq
2!
www.cgaspirants.com
www.cgaspirants.com
t3
μ3 = coefficient of = npq (q – p)
3!
t4
μ4 = coefficient of = npq (1 – 3 pq) + 3n(n – 1) p2q2
4!
= 3n2p2q2 + npq (1 – 6pq)
Hence,
μ 23 (q − p) 2 (1 − 2 p) 2
β1 = = =
μ 32 npq npq
1 − 2p
∴ γ1 =
m
npq
μ4 1 − 6 pq
and β2 = =3+
o
μ 22 npq
.c
1 − 6 pq
∴ γ2 =
npq
Note 1. γ1 =
1 − 2p
positive, if p >
npq
1
n
, skewness is negative and if p = 1
ts
gives a measure of skewness of the binomial distribution. If p <
, it is zero.
1
2
, skewness is
2 2
ra
1 − 6 pq
β2 = 3 + gives a measure of the kurtosis of the binomial distribution.
npq
pi
Note 2. If n independent trials constitute one experiment and this experiment is repeated N times then
the frequency of r successes is N . nCr pr qn–r.
as
3.55 APPLICATIONS OF BINOMIAL DISTRIBUTION
1. In problems concerning no. of defectives in a sample production line.

g
2. In estimation of reliability of systems.

.c
3. No. of rounds fired from a gun hitting a target.

4. In Radar detection.
w
w
EXAMPLES
w
Example 1. (i) Comment on the following statement:

For a Binomial distribution, mean is 6 and variance is 9.
(ii) A die is tossed thrice. A success is getting 1 or 6 on a toss. Find the mean and variance
of the number of success.
Sol. (i) μ = np = 6 ...(1)
2
σ = npq = 9 ...(2)
Dividing (2) by (1), we get
9
q = = 1.5
6
which is impossible as 0 ≤ q ≤1
∴ The above statement is False.
www.cgaspirants.com
www.cgaspirants.com
2 1
(ii) Prob. of getting success (1 or 6) on a toss = = =p
6 3
1 2
∴ q=1– =
3 3
No. of tosses of a die, n = 3
FG 1 IJ = 1. FG 1 IJ FG 2 IJ = 2 .
(i) Mean = np = 3
H 3K (ii) Variance = npq = (3)
H 3K H 3K 3
Example 2. If 10% of the bolts produced by a machine are defective, determine the
probability that out of 10 bolts choosen at random
(i) 1 (ii) None (iii) at most 2 bolts will be defective.
10 1
m
Sol. Here, p(defective) = = (given)
100 10
1 9
o
∴ q(non-defective) = 1 – =
10 10
.c
Also, n = 10, (n is no. of bolts chosen). (given)
The probability of r defective bolts out of n bolts chosen at random is given by
(i) Here r = 1,
P(r) = nCr pr qn–r
n ts ...(1)
FG 1 IJ FG 9 IJ
1 10 − 1
ra
P(1) = 10C1 | Using (1)
∴
H 10 K H 10 K
F1IF 9I 9
pi
= 10 G J G J = (.9) 9 = 0.3874 ...(2)

H 10 K H 10 K
(ii) Here r = 0
as
P(0) = 10C0
FG 1 IJ FG 9 IJ
0 10 − 0
=
FG 9 IJ 10
= 0.3486 ...(3) | Using (1)
∴
H 10 K H 10 K H 10 K
g
.c
(iii) Prob. that at most 2 bolts will be defective = P(r ≤ 2) = P(0) + P(1) + P(2) ...(4)
FG 1 IJ FG 9 IJ
2 10 − 2
w
Now, P(2) = 10C2 | Using (1)

H 10 K H 10 K
F 1 IJ (0.43046) = 0.1937
w
= 45 G
H 100 K
w
∴ From (4), Required Probability = P(0) + P(1) + P(2)

= 0.3486 + 0.3874 + 0.1937 = 0.9297.
Example 3. A binomial variable X satisfies the relation 9P (X = 4) = P(X = 2) when n = 6.
Find the value of the parameter p and P(X = 1).
Sol. We know that
P(X = r) = nCr pr qn–r ...(1)
∴ P(X = 4) = 6C p4 q2 = 15p4q2
4
and P(X = 2) = 6 C p 2 q4 = 15 p2 q4 | Since n = 6
2
The given relation is
9 P(X = 4) = P(X = 2) ⇒ 9(15p4 q2) = 15p2q4
www.cgaspirants.com
www.cgaspirants.com
⇒ 9p2 = q2 = (1 – p)2 |∵ p+q=1

⇒ 9p2 =1+ p2 – 2p
⇒ 8p2 + 2p – 1 = 0 ⇒ (4p – 1) (2p + 1) = 0
1
∴ p= |∵ p cannot be negative
4
Now, P(X = 1) = 6C1

FG 1 IJ FG 3 IJ
1 5
= .3559. ∵ q=1− p=
3
H 4K H 4K 4
Example 4. Fit a binomial distribution to the following frequency data:
x: 0 1 2 3 4
m
f: 30 62 46 10 2
Sol. The table is as follows:
o
.c
x f fx
0 30 0
1
2
3
n 62
46
10
ts 62
92
30
ra
4 2 8
Σf = 150 Σfx = 192
pi
Σ fx 192
Mean of observations = = = 1.28
Σ f 150
as
⇒ np = 1.28
⇒ 4p = 1.28 (n is no. of trials)
g
⇒ p = 0.32
.c
∴ q = 1 – p = 1 – 0.32 = 0.68
Also, N = 150 | ∵ N = Σf
w
Hence the binomial distribution is = N(q + p)n = 150 (0.68 + 0.32)4.

w
Example 5. A student is given a true-false examination with 8 questions. If he corrects at

least 7 questions, he passes the examination. Find the probability that he will pass given that he
guesses all questions.
w
Sol. Here, n = no. of questions asked = 8

1 1
p= ,q= | Since the question can either be true or false
2 2
Probability that he will pass
= P(r ≥ 7) = P(7) + P(8)
= 8 C7 FG 1 IJ FG 1 IJ
7 8 −7
FG IJ FG 1 IJ
+ 8 C8
1
8 8− 8
FG 1 IJ FG 1 IJ
7 1
FG 1 IJ 8
H 2K H 2K H K H 2K
2
=8
H 2K H 2K +1.
H 2K
=
FG 1 IJ 8
9
= .03516.
H 2K (8 + 1) =
256
www.cgaspirants.com
www.cgaspirants.com
Example 6. During war, 1 ship out of 9 was sunk on an average in making a certain
voyage. What was the probability that exactly 3 out of a convoy of 6 ships would arrive safely?
1 8 1
Sol. p, the probability of a ship arriving safely = 1 – = ; q= ,n=6
9 9 9
The probability that exactly 3 ships arrive safely = P(r = 3) = 6C3

10240
. FG 1IJ FG 8 IJ
3 3
96 H 9K H 9K =
Example 7. A policeman fires 6 bullets on a dacoit. The probability that the dacoit will be
killed by a bullet is 0.6. What is the probability that dacoit is still alive?
Sol. Here n = no. of bullets fired = 6, p = 0.6, q = 1 – p = 0.4
Probability that dacoit is still alive (not killed)
m
= P(r = 0) = nC0 p0 qn–0 = 6C0 (.6)0 (.4)6 = (.4)6 = .004096.
Example 8. If the probability of hitting a target is 10% and 10 shots are fired indepen-
o
dently. What is the probability that the target will be hit at least once?
10 1 1 9
.c
Sol. Here, p = = ,q=1–p=1– = , n = 10
100 10 10 10
Probability that the target will be hit at least once
= P(r ≥ 1) = 1 – P(r = 0)
n
= 1 – [nC0 p0 qn] = 1 – 10 C0
LM ts FG 1 IJ FG 9 IJ OP = 0.6513.
0 10
MN H 10 K H 10 K QP
ra
Example 9. Out of 800 families with 4 children each, how many families would be expected
pi
to have (i) 2 boys and 2 girls (ii) at least one boy (iii) no girl (iv) atmost two girls? Assume equal
probabilities for boys and girls. (U.P.T.U. 2014)
Sol. Since probabilities for boys and girls are equal,
as
1 1
p = probability of having a boy = ; q = probability of having a girl =
2 2
g
n=4 N = 800
.c
(i) The expected number of families having 2 boys and 2 girls

FG 1 IJ FG 1 IJ
2 2
w
1
= 800 4C2 = 800 × 6 ×
= 300.
H 2K H 2K 16
w
(ii) The expected number of families having at least one boy

LM
1 FG IJ FG 1 IJ +
3
FG 1 IJ FG 1 IJ
2 2
FG 1 IJ FG 1 IJ 3
FG 1 IJ 4 OP
w
= 800 4 C1 4
C2 + 4 C3 + 4 C4
MN
2 H K H 2K H 2K H 2K H 2K H 2K H 2K PQ
1
= 800 × [4 + 6 + 4 + 1] = 750
16
(iii) The expected number of families having no girl i.e., having 4 boys
= 800 . 4C4
FG 1 IJ 4
= 50.
H 2K
(iv) The expected number of families having atmost two girls i.e., having at least 2 boy
LM
= 800 4 C2
1 FG IJ FG 1 IJ
2 2
+ 4 C3
FG 1 IJ FG 1 IJ 3
+ 4 C4
FG 1 IJ 4 OP = 800 × 1 [6 + 4 + 1] = 550.
MN
2 H K H 2K H 2K H 2K H 2K PQ 16
www.cgaspirants.com
www.cgaspirants.com
Example 10. Six dice are thrown 729 times. How many times do you expect at least
three dice to show a five or six?
2 1
Sol. p = the chance of getting 5 or 6 with one die = =
6 3
1 2
q = 1 – = , n = 6, N = 729
3 3
since dice are in sets of 6 and there are 729 sets.
The expected number of times at least three dice showing five or six
= N . P(r ≥ 3)
= 729 [P(3) + P(4) + P(5) + P(6)]
LM FG 2 IJ FG 1 IJ FG 2 IJ FG 1 IJ FG 2 IJ FG 1 IJ FG 1 IJ OP
m
3 3 2 4 5 6
6
C3 + 6 C4 + 6 C5 + 6 C6
= 729
MN H 3K H 3K H 3K H 3K H 3K H 3K H 3K PQ
o
729
= [160 + 60 + 12 + 1] = 233.
.c
36
1
Example 11. The probability of a man hitting a target is
1
n 3
. How many times must he
fire so that the probability of his hitting the target at least once is more than 90%? ts
ra
Sol. p=
3
The probability of not hitting the target in n trials is qn.
pi
Therefore, to find the smallest n for which the probability of hitting at least once is more
than 90%, we have
1 – qn > 0.9
as
1–
FG 2 IJ n
> 0.9
⇒
H 3K
g
.c
FG 2 IJ n
< 0.1
⇒
H 3K
w
The smallest n for which the above inequality holds true is 6 hence he must fire 6 times.
w
Example 12. In a bombing action, there is 50% chance that any bomb will strike the
target. Two direct hits are needed to destroy the target completely. How many bombs are re-
w
quired to be dropped to give a 99% chance or better of completely destroying the target?
50 1
Sol. We have, p = =
100 2
Since the probability must be greater than 0.99, if n bombs are dropped, we have
nC FG 1 IJ n
+ n C3
FG 1 IJn
+ n C4
FG 1 IJ n
+ ...... + nCn
FG 1 IJ n
≥ 0.99
2
H 2K H 2K H 2K H 2K
FG 1 IJ n
[nC2 + nC3 + nC4 + ...... + nCn] ≥ 0.99
⇒
H 2K
2n − n − 1
≥ 0.99
2n
www.cgaspirants.com
www.cgaspirants.com
1+n
⇒ 1− ≥ 0 .99
2n
1+n
⇒ ≤ 0.01
2n
⇒ 2n ≥ 100n + 100
By trial, n = 11 satisfies the inequality.
Hence 11 bombs are required to be dropped.
ASSIGNMENT
m
1. (i) Ten coins are tossed simultaneously. Find the probability of getting at least seven heads.
(ii) A die is thrown five times. If getting an odd number is a success, find the probability of getting
o
at least four successes. (M.T.U. 2012)
2. (a) The probability of any ship of a company being destroyed on a certain voyage is 0.02. The
.c
company owns 6 ships for the voyage. What is the probability of :
(i) losing one ship (ii) losing atmost two ships (iii) losing none?
ts
(b) Assume that on the average one telephone number out of fifteen called between 2 P.M. and
3 P.M. on week-days is busy. What is the probability that if 6 randomly selected telephone
n
numbers are called (i) not more than 3 (ii) at least 3 of them will be busy?
3. (i) The incidence of occupational disease in an industry is such that the workers have a 20%
ra
chance of suffering from it. What is the probability that out of six workers chosen at random,
four or more will suffer from the disease?
pi
(ii) The probability that a man aged 60 will live to be 70 is 0.65. What is the probability that out
of 10 men, now 60, at least 7 will live to be 70?
3
as
4. (i) If the mean of a binomial distribution is 3 and the variance is 2

, find the probability of
obtaining at least 4 successes.
(ii) In a binomial distribution, for n = 5 if P(x = 1) = 0.4096 and P(x = 2) = 0.2048, then find the
g
value of p.
.c
25 50
(iii) The sum and product of the mean and variance of a binomial distribution are and
3 3
w
respectively. Find the distribution. (U.P.T.U. 2007)

(iv) If the probability of a defective bolt is 0.1, find (a) The mean (b) The standard deviation for the
w
distribution in a total of 400 bolts.
FG 1 + 2 et IJ 5 , find P(X = 2).

w
(v) If the moment generating function of a random variable X is

H3 3 K
1
5. (a) The probability that a bomb dropped from a plane will strike the target is . If six bombs are
5
dropped, find the probability that (i) exactly two will strike the target, (ii) at least two will strike
the target.
(b) Four persons in a group of 20 are graduates. If 4 persons are selected at random from 20, find
the probability that
(i) all are graduates (ii) at least one is a graduate.
6. A bag contains 5 white, 7 red and 8 black balls. If four balls are drawn, one by one, with replace-
ment, what is the probability that
(i) none is white (ii) all are white
(iii) at least one is white (iv) only 2 are white?
www.cgaspirants.com
www.cgaspirants.com
7. (i) In a hurdle race, a player has to cross 10 hurdles. The probability that he will clear each hurdle
is 56 . What is the probability that he will knock down fewer than 2 hurdles ?
(ii) If on an average one ship in every ten is wrecked, find the probability that out of 5 ships
expected to arrive, 4 at least will arrive safely.
8. The prob. that a bulb produced by a factory will fuse after a use of 150 days is 0.05. Find the prob.
that out of 5 such bulbs [M.T.U. (B. Pharma) 2011]
(i) None (ii) Atmost one
(iii) More than one (iv) At least one will fuse after 150 days of use.
9. Five cards are drawn successively with replacement from a well-shuffled deck of 52 cards. What
is the probability that
(i) all the five cards are spades (ii) only three are spades (iii) none is spade?
m
10. Manish takes a step forward with probability 0.4 and backward with probability 0.6. Find the
probability that at the end of 11 steps, he is one step away from the starting point.
o
[Hint. Manish will take either 6 steps forward and 5 backward or 5 steps forward and 6 backward].
11. (a) In 800 families with 5 children each, how many families would be expected to have (i) 3 boys
.c
and 2 girls, (ii) 2 boys and 3 girls, (iii) no girl (iv) at the most two girls. (Assume probabilities
for boys and girls to be equal.)
ts
(b) Out of 800 families with 5 children each, how many would you expect to have (i) 3 boys
(ii) 5 girls (iii) either 2 or 3 boys? Assume equal probabilities for boys and girls.
n
(M.T.U. 2012)
ra
(c) Out of 320 families with 5 children each, what percentage would be expected to have (i) 2 boys
and 3 girls (ii) at least one boy? Assume equal probability for boys and girls.
(G.B.T.U. 2011)
pi
12. In 100 sets of ten tosses of an unbiased coin, in how many cases do you expect to get
(i) 7 heads and 3 tails (ii) at least 7 heads?
as
13. The following data are the number of seeds germinating out of 10 on damp filter for 80 sets of
seeds. Fit a binomial distribution to this data:
g
x: 0 1 2 3 4 5 6 7 8 9 10 Total
.c
f: 6 20 28 12 8 6 0 0 0 0 0 80
Σfx
w
[Hint. Here n = 10, N = 80, Mean = = 2.175 ∴ np = 2.175 etc.]

Σf
14. Fit a binomial distribution for the following data and compare the theoretical frequencies with
w
the actual ones :

x: 0 1 2 3 4 5
w
f: 2 14 20 34 22 8
15. Fit a binomial distribution to the data given in the following table:
(i) x : 0 1 2 3 4
f: 24 41 28 5 2 (M.T.U. 2012)
(ii) x : 0 1 3 4
f: 28 62 10 4 (U.K.T.U. 2011)
16. (i) Assuming half the population of a town consumes chocolates so that the chance of an individual
1
being consumer is and that 100 investigators each take 10 individuals to see whether they
2
are consumers, how many investigators would you expect to report that 3 people or less were
consumers? (U.P.T.U. 2006)
www.cgaspirants.com
www.cgaspirants.com
(ii) Assuming that 20% of the population of a city are literate, so that the chance of an individual
1
being literate is and assuming that 100 investigators each take 10 individuals to see
5
whether they are literate, how many investigators would you expect to report 3 or less were
literate?
17. Following results were obtained when 100 batches of seeds were allowed to germinate on damp
1 89
filter paper in a laboratory : β1 = , β2 = . Determine the Binomial distribution. Calculate the
15 30
expected frequency for x = 8 assuming p > q.
18. A coffee connoisseur claims that he can distinguish between a cup of instant coffee and a cup of
percolator coffee 75% of the time. It is agreed that his claim will be accepted if he correctly
m
identifies at least 5 of the 6 cups. Find his chances of having the claim (i) accepted (ii) rejected
when he does have the ability he claims.
19. A multiple-choice test consists of 8 questions with 3 answers to each question of which only one is
o
correct. A student answers each question by rolling a balanced die and checking the first answer
.c
if he gets 1 or 2, the second answer if he gets 3 or 4 and the third answer if he gets 5 or 6. To get a
distinction, the student must secure at least 75% correct answers. If there is no negative marking,
20.
ts
what is the probability that the student secures a distinction?
An irregular six-faced die is thrown and the expectation that in 10 throws, it will give five even
n
numbers is twice the expectation that it will give four even numbers. How many times in 10,000
sets of 10 throws each, would you expect it to give no even number?
ra
Answers
pi
11 3
1. (i) (ii)
64 16
2. (a) (i) 0.1085, (ii) 0.9997, (iii) 0.8858 (b) (i) 0.9997 (ii) 0.005
as
53
3. (i) (ii) 0.5137
3125
g
11 1 FG 2 + 1 IJ 15
.c
4. (i) , (ii) , (iii)

32 5 H 3 3K
(iv) (a) 40 (b) 6 (v) 0.1646
w
5. (a) (i) 0.246 (ii) 0.345 (b) (i) 0.0016 (ii) 0.5904
FG IJ , 9
w
81 1 175 27 5 5
6. (i)
256
, (ii)
256
, (iii)
256
, (iv)
128
7. (i)
H K
2 6
(ii) 0.91854
w
8. (i)
FG 19 IJ 5 (ii)
FG IJ 4
6 19 6 F 19 I
(iii) 1 – G J
4
(iv) 1 –
FG 19 IJ 5
H 20 K H K
5 20 5 H 20 K H 20 K
9.
F 1I5
(i) G J
F 1I5
(ii) 90 G J
F 3I5
(iii) G J 10. 0.36787
H 4K H 4K H 4K
11. (a) (i) 250 (ii) 250 (iii) 25 (iv) 400
(b) (i) 250 (ii) 25 (iii) 500
(c) (i) 31.25% (ii) 96.875%
12. (i) 12 nearly (ii) 17 nearly 13. 80 (0.7825 + 0.2175)10
14. 100 (0.432 + 0.568)5 15. (i) 100 (0.7 + 0.3)4, (ii) 104 (0.7404 + 0.2596)4
www.cgaspirants.com
www.cgaspirants.com
16. (i) 17 (ii) 88 17. 100 FG 1 + 3 IJ 20

, 0.075168752
H 4 4K
18. (i) 0.534 (ii) 0.466
19. 0.0197 20. 1.
POISSON DISTRIBUTION
3.56 POISSON DISTRIBUTION AS A LIMITING CASE OF BINOMIAL DISTRIBUTION

[U.P.T.U. 2007 ; G.B.T.U. 2010, 2013 ; M.T.U. 2013, 2014]
m
Poisson distribution was discovered by S.D. Poisson in the year 1837.
o
If the parameters n and p of a binomial distribution are known, we can find the distribution.
But in situations where n is very large and p is very small, application of binomial distribution
.c
is very labourious. However, if we assume that as n → ∞ and p → 0 such that np always remains
finite, say λ, we get the Poisson approximation to the binomial distribution.
Now, for a binomial distribution
P(X = r) = nCrqn–rpr
n(n − 1)(n − 2) ... (n − r + 1)
n ts
= × (1 – p)n–r × pr
ra
r!
n(n − 1)(n − 2) ... (n − r + 1) λ FG IJ n− r
FG λ IJ r
λ
= × 1−
H K ×
H nK | Since np = λ ∴ p=
pi
r! n n
FG 1 − λ IJ n
n(n − 1)(n − 2) ... (n − r + 1) H nK

as
r
λ
= × ×
r! n r
FG 1 − λ IJ r
H nK
g
FG 1 − λ IJ
.c
λ F n I F n − 1I F n − 2 I
r
F n − r + 1I H nK
G J G J G J ...... G J
w
=
H
r! n K H n K H n K H n K ×
FG 1 − λ IJ r
H nK
w
LMF λ I OP −
n −λ
GH 1 − n JK P
w
λ
M
λ F
G 1 − n1 IJK FGH 1 − n2 IJK ...... FGH 1 − r n− 1IJK × MN F λ I PQ
r
r! H
= r
GH 1 − n JK
As n → ∞, each of the (r – 1) factors
FG 1 − 1 IJ , FG 1 − 2 IJ , ......, FG 1 − r − 1IJ FG λ IJ r
H nK H nK H nK tends to 1. Also 1 −
H n K tends to 1.
F 1I x LMF λ I − OP
n −λ
Lt G 1 + J = e, the Naperian base. MMGH 1 − n JK

λ
Since
H xK ∴ PP ⎯→ e–λ as n → ∞
x→±∞
N Q
www.cgaspirants.com
www.cgaspirants.com
Hence in the limiting case when n → ∞, we have

−λ r
P(X = r) = e . λ (r = 0, 1, 2, 3, ...) ...(1)
r!
where λ is a finite number = np.

(1) represents a probability distribution which is called the Poisson probability distribution.
Note 1. λ is called the parameter of the distribution.
Note 2. The sum of the probabilities P(r) for r = 0, 1, 2, 3, ...... is 1, since
λe− λ λ2 e− λ λ3 e− λ
P(0) + P(1) + P(2) + P(3) + ...... = e–λ + + + + ......
1! 2! 3!
F I
m
λ λ2 λ3
= e–λ 1 +GH + +
1! 2 ! 3 ! JK
+ ...... = e–λ . eλ = 1.
o
3.57 RECURRENCE FORMULA FOR THE POISSON DISTRIBUTION
.c
[U.P.T.U. (B. Pharma.) 2009]
For Poisson distribution,

e −λ λr
P(r) =
r!
and P(r + 1) =
n e − λ λr +1
(r + 1) !
ts
P(r + 1) λr ! λ
ra
∴ = =
P ( r) (r + 1) ! r + 1
λ
pi
or P(r + 1) = P(r), r = 0, 1, 2, 3, ......

r+1
This is called the recurrence or recursion formula for the Poisson distribution.
as
3.58 MEAN AND VARIANCE OF THE POISSON DISTRIBUTION [U.P.T.U. 2006]

g
e −λ λr
For the Poisson distribution, P(r) =
.c
r!
∞ ∞
w
e − λ λr
Mean μ = ∑ rP(r) = ∑ r.
r!
r=0 r=0
w
∞
λr F λ2 λ3 I
= e–λ ∑ (r − 1) ! = e GH λ + 1 ! + 2 ! + ......JK
−λ
w
r=1
F λ λ2 I
= λe–λ 1 + GH +
1! 2 ! JK
+ ...... = λe–λ . eλ = λ
Thus, the mean of the Poisson distribution is equal to the parameter λ.

∞ ∞ ∞
λr e − λ r 2 λr
Variance σ2 = ∑
r=0
r 2 P(r) – μ 2 = ∑
r=0
r2 .
r!
− λ2 = e − λ ∑
r=1
r!
– λ2
= e–λ
LM1 . λ + 2 . λ
2 2 2
+
3 2 . λ3 4 2 . λ4
+ + ...... – λ2
OP
MN 1 ! 2 ! 3! 4! PQ
www.cgaspirants.com
www.cgaspirants.com
= λe–λ 1 +
LM 2λ + 3λ + 4λ + ......OP – λ
2 2 3
2
MN 1 ! 2 ! 3 ! PQ
= λe–λ
LM1 + (1 + 1)λ + (1 + 2)λ + (1 + 3)λ + ......OP – λ
2 3
2
N 1! 2! 3! Q
LMF 1 + λ + λ + λ + ......I + F λ + 2λ + 3λ + ......I OP – λ
2 3 2 3
= λe–λ
MNGH 1 ! 2 ! 3 ! JK GH 1 ! 2 ! 3 ! JK PQ 2
LMe + λ F 1 + λ + λ + ......I OP
2
MN GH 1 ! 2 ! JK PQ – λ
λ
= λe–λ 2
m
= λe–λ [eλ + λeλ] – λ2 = λe–λ . eλ (1 + λ) – λ2 = λ(1 + λ) – λ2 = λ.
Hence, the variance of the Poisson distribution is also λ.
o
Thus, the mean and the variance of the Poisson distribution are each equal to
.c
the parameter λ.
Note 1. The mean and the variance of the Poisson distribution can also be derived from those of the
ts
binomial distribution in the limiting case when n → ∞, p → 0 and np = λ.
Mean of Binomial distribution is np.
n
∴ Mean of Poisson distribution = Lt np = Lt λ = λ
n→∞ n →∞
ra
Variance of Binomial distribution is npq = np (1 – p)
Variance of Poisson distribution = Lt np (1 – p) = Lt λ 1 −

FG λIJ
∴
H K
= λ.
pi
n→∞ n→∞ n
Note 2. For Poisson distribution, μ3 = λ and μ4 = 3λ2 + λ.
as
Coefficients of skewness and kurtosis are given by

1 1 1 1
β1 =and γ1 = . Also, β2 = 3 + and γ2 =
g
λ λ λ λ
Hence Poisson distribution is always a skewed distribution.
.c
Remark. While fitting the Poisson distribution to a given data, we round the figures to the nearest
integer but it should be kept in mind that the total of the observed and the expected frequencies should
w
be same.
w
3.59 MODE OF POISSON DISTRIBUTION

w
λr
Let P(x = r) = e–λ , r = 0, 1, 2, ......, ∞
r!
The value of r which has a greater probability than any other value is the mode of the
Poisson distribution. Thus r is mode if
P(X = r) ≥ P(X = r + 1) and P(X = r) ≥ P(X = r – 1)
e − λ . λr e − λ . λr + 1 e − λ . λr e − λ . λr − 1
⇒ ≥ and ≥
r! ( r + 1) ! r! ( r − 1) !
λ λ
⇒ 1≥ and ≥1
r +1 r
⇒ r ≥ λ – 1 and r ≤ λ i.e., λ – 1 ≤ r ≤ λ
www.cgaspirants.com
www.cgaspirants.com
Case I. If λ is a positive integer, there are two modes λ – 1 and λ.

Case II. If λ is not a positive integer, there is one mode and is the integral value be-
tween λ – 1 and λ.
3.60 APPLICATIONS OF POISSON DISTRIBUTION
This distribution is applied to problems concerning :

(i) Arrival pattern of defective vehicles in a workshop.
(ii) Patients in a hospitals.
(iii) Telephone calls.
(iv) Demand pattern for certain spare parts.
m
(v) Number of fragments from a shell hitting a target.
(vi) Emission of radioactive (α) particles.
o
.c
EXAMPLES
ts
Example 1. If the variance of the Poisson distribution is 2, find the probabilities for
r = 1, 2, 3, 4 from the recurrence relation of the Poisson distribution. Also find P(r ≥ 4).
n
(M.T.U. 2013)
Sol. λ, the parameter of Poisson distribution = Variance = 2
ra
Recurrence relation for the Poisson distribution is

λ 2
pi
P(r + 1) = P(r) = P(r) ...(1)

r+1 r+1
e −λ λr e −2 (2) 0
as
Now P(r) = ⇒ P(0) = = e–2 = 0.1353

r! 0!
Putting r = 0, 1, 2, 3 in (1), we get
g
2
.c
P(1) = 2P(0) = 2 × 0.1353 = 0.2706 ; P(2) = P(1) = 0.2706

2
2 2 2 1
w
P(3) = P(2) = × 0.2706 = 0.1804 ; P(4) = P(3) = × 0.1804 = 0.0902.

3 3 4 2
Now, P(r ≥ 4) = 1 – [P(0) + P(1) + P(2) + P(3)]
w
= 1 – [0.1353 + 0.2706 + 0.2706 + 0.1804] = 0.1431.

w
Example 2. Using Poisson distribution, find the probability that the ace of spades will be
drawn from a pack of well-shuffled cards at least once in 104 consecutive trials. (U.P.T.U. 2015)
1
Sol. p= , n = 104
52
1
∴ λ = np = 104 × =2
52
Prob. (at least once) = P(r ≥ 1) = 1 – P(0)
e −λ . λ0
=1– = 1 – e–2 = 1 – 0.135335 ≈ 0.8647.
0!
www.cgaspirants.com
www.cgaspirants.com
Example 3. (i) Fit a Poisson distribution to the following data and calculate theoretical
frequencies.
Deaths: 0 1 2 3 4
Frequencies: 122 260 15 2 1
[U.P.T.U. 2014 ; U.K.T.U. 2010]
(ii) The frequency of accidents per shift in a factory is shown in the following table:
Accident per shift Frequency

0 192
1 100
2 24
m
3 3
4 1
o
Total 320
.c
Calculate the mean number of accidents per shift. Fit a Poisson distribution and calcu-
late theoretical frequencies.
Sol. (i) Mean of given distribution =

n Σ fx
Σf
ts
ra
60 + 30 + 6 + 4
⇒ λ= = 0.5
200
pi
e − λ . λr e − .5 (.5) r (0.5) r
Required Poisson distribution = N . = 200 . = (121.306)
r! r! r!
as
r N. P(r) Theoretical frequency

g
(0.5) 0
0 121.306 × = 121.306 121
.c
0!
(0.5)1
w
1 121.306 × = 60.653 61
1!
w
(0.5) 2
2 121.306 × = 15.163 15
2!
w
(0.5) 3
3 121.306 × = 2.527 3
3!
(0.5) 4
4 121.306 × = 0.3159 0
4!
Total = 200
(ii) Mean number of accidents per shift

Σ fx 100 + 48 + 9 + 4
λ= = = 0.5031
Σ f 320
www.cgaspirants.com
www.cgaspirants.com
∴ Required Poisson distribution

e − λ . λr e −.5031 (.5031) r (193.48)(.5031) r
=N. = 320 . =
r! r! r!
r N. P(r) Theoretical frequency
0 193.48 194
1 97.34 97
2 24.38 24
3 4.10 4
4 0.51 1
m
Total = 320
Example 4. (i) Suppose that a book of 600 pages contains 40 printing mistakes. Assume
o
that these errors are randomly distributed throughout the book and r, the number of errors per
page has a Poisson distribution. What is the probability that 10 pages selected at random will be
.c
free from errors?
(ii) Wireless sets are manufactured with 25 solders joints each, on the average 1 joint in
40
n
1
ts
500 is defective. How many sets can be expected to be free from defective joints in a consignment
of 10000 sets?
Sol. (i) p= , n = 10
ra
=
600 15
∴ λ = np = 10
1 FG IJ
2
H K
=
pi
15 3
e − λ λr e − 2 / 3 (2/3) r
as
P(r) = =
r! r!
e − 2 / 3 (2/3) 0
= e–2/3 = 0.51.
g
∴ P(0) =
0!
.c
1
(ii) p= , n = 25
500
w
1 1
∴ λ = np = 25 × = = 0.05
w
500 20
No. of sets in a consignment, N= 10000
w
e − .05 (0.05) 0
Probability of having no defective joint = P(r = 0) = = 0.9512.
0!
∴ The expected no. of sets free from defective joints = 0.9512 × 10000 = 9512.
Example 5. A manufacturer knows that the condensors he makes contain on an average
1% of defectives. He packes them in boxes of 100. What is the probability that a box picked at
random will contain 4 or more faulty condensors?
Sol. p = 0.01, n = 100
∴ λ = np = 1
e − λ λr e − 1
P(r) = =
r! r!
www.cgaspirants.com
www.cgaspirants.com
P(4 or more faulty condensors) = P(4) + P(5) + ... + P(100)

= 1 – [P(0) + P(1) + P(2) + P(3)]
=1–
LM e + e + e + e OP
−1 −1 −1 −1
MN 0 ! 1 ! 2 ! 3 ! PQ
L 1 1O
= 1 – e M1 + 1 + + P = 1 −
–1 8
= 1 – 0.981 = 0.019.
N 2 6Q 3e
Example 6. (i) If the probabilities of a bad reaction from a certain injection is 0.0002,
determine the chance that out of 1000 individuals more than two will get a bad reaction.
(ii) The probability that a man aged 50 years will die within a year is 0.01125. What is the
probability that of 12 such men, at least 11 will reach their 51st birthday?
m
(Given: e–.135 = 0.87366)
Sol. (i) Here, p = 0.0002, n = 1000
o
∴ λ = np = 1000 × 0.0002 = 0.2.
.c
Since the prob. of bad reaction is very small and no. of trials is very high, we use Poisson
distribution here.
n
e −0.2 (0.2) 0
ts
The prob. that out of 100 individuals, more than 2 will get a bad reaction is
= P(r > 2) = 1 – P(r ≤ 2) = 1 – [P(0) + P(1) + P(2)] ...(1)
ra
Now, P(0) = = 0.8187 (Here r = 0)
0
e −0.2 (0.2)1
pi
P(1) = = 0.1637 (Here r = 1)

1
as
e −0.2 (0.2) 2
and P(2) = = 0.0164. (Here r = 2)
2
∴ From (1), Reqd. probability = 1 – [0.8187 + 0.1637 + 0.0164] = 0.0012.
g
(ii) p = 0.01125, n = 12
.c
∴ λ = np = 12 × 0.01125 = 0.135
w
P (at least 11 survive) = P(atmost 1 dies)
e − λ . λ0 e − λ . λ1
w
= P(0) + P(1) = +
0! 1!
w
= e–.135 (1 + 0.135) = 1.135 × 0.87366 = 0.9916.

Example 7. A car-hire firm has two cars, which it hires out day by day. The number of
demands for a car on each day is distributed as a Poisson distribution with mean 1.5. Calculate
the proportion of days on which neither car is used and the proportion of days on which some
demand is refused (e–1.5 = 0.2231).
Sol. Since the number of demands for a car is distributed as a Poisson distribution with
mean λ = 1.5.
∴ Proportion of days on which neither car is used
= Probability of there being no demand for the car
λ0 e − λ
= = e–1.5 = 0.2231
0!
www.cgaspirants.com
www.cgaspirants.com
Proportion of days on which some demand is refused

= probability for the number of demands to be more than two
F λe − λ λ2 e − λ I
GH
= 1 – P (x ≤ 2) = 1 – e − λ +
1!
+
2! JK
F1 + 1.5 + (1.5) I 2
=1– e–1.5 GH 2 JK
= 0.1912625.
Example 8. Suppose the number of telephone calls on an operator received from 9 : 00 to

9 : 05 follow a Poisson distribution with a mean 3. Find the probability that
(i) The operator will receive no calls in that time interval tomorrow.
m
(ii) In the next three days, the operator will receive a total of 1 call in that time interval.
(Given: e–3 = 0.04978)
o
Sol. Here, λ = 3
.c
e −λ . λ0
(i) P(0) = = e–3 = 0.04978
0!
(ii)
|R e .λ |UV
−λ 0 2
e − λ . λ1
ts
Reqd. probability = P(0) P(0) P(1) + P(0) P(1) P(0) + P(1) P(0) P(0)
n
= 3S = 9(e–3)3 = 0.00111.
|T 0 ! |W
ra
1!
Example 9. The no. of arrivals of customers during any day follows Poisson distribution
pi
with a mean of 5. What is the probability that the total no. of customers on two days selected at
random is less than 2? (Given: e–10 = 4.54 × 10–5)
as
Sol. λ=5
Arrival of Customers
g
I day II day Total

.c
0 0 0
0 1 1
w
1 0 1
w
Reqd. probability = P(0) P(0) + P(0) P(1) + P(1) P(0)

e −5 .5 0 e −5 .5 0 e −5 .5 0 e −5 .51 e −5 .51 e −5 .5 0
w
= . + . + .
0! 0! 0! 1! 1! 0!
= e–10 + 2 . 5 . e–10 = 11 e–10 = 11 × 4.54 × 10–5
= 4.994 × 10–4.
Example 10. An insurance company finds that 0.005% of the population dies from a
certain kind of accident each year. What is the probability that the company must pay off no
more than 3 of 10,000 insured risks against such incident in a given year?
0.005
Sol. p= = 0.00005, n = 10000
100
∴ λ = np = 10000 × 0.00005 = 0.5
www.cgaspirants.com
www.cgaspirants.com
Reqd. Probability = 1 – P(r ≤ 3) = 1 – [P(0) + P(1) + P(2) + P(3)]
=1–
LM e − 0.5
(0.5) 0 e − 0.5 (0.5)1 e − 0.5 (0.5) 2 e − 0.5 (0.5) 3
+ + +
OP
MN 0! 1! 2! 3! PQ
= 1 – e–5 [1 + 0.5 + 0.125 + 0.021] = 0.0016.
Example 11. In a certain factory turning out razor blades, there is a small chance of
0.002 for any blade to be defective. The blades are supplied in packets of 10. Calculate the
approximate number of packets containing no defective, one defective and two defective blades
in a consignment of 10,000 packets. (Given: e–0.02 = 0.9802) [U.P.T.U. 2009]
Sol. p(defective) = 0.002
n = 10 (no. of blades in a packet)
m
∴ λ = np = 10 × 0.002 = 0.02
No. of packets in the consignment, N = 10,000.
o
e −0.02 (0.02) 0
.c
(i) Probability of having no defective = P(0) = = 0.9802 | Here r = 0
0!
n ts
Approximate no. of packets having zero defective in the consignment = 0.9802 × 10000
= 9802
e −0.02 (0.02)1
ra
(ii) Probability of having one defective = P(1) = = 0.9802 × 0.02 = 0.019604
1!
Approximate no. of packets having one defective in the consignment
pi
= 0.019604 × 10000 ≈ 196.

(iii) Probability of having two defective blades
as
e −0.02 (0.02) 2 (0.980198) × (0.0004)

P(2) = = = 0.000196.
2! 2
g
∴ Approximate no. of packet having two defectives in the consignment

.c
= 0.000196 × 10000 = 1.96 ≈ 2.

w
Example 12. (i) Six coins are tossed 6400 times. Using the Poisson distribution, deter-
mine the approximate probability of getting six heads x times. [U.P.T.U. (C.O.) 2008]
w
(ii) A Poisson distribution has a double mode at x = 3 and x = 4. What is the probability
that x will have one or the other of these two values? [U.P.T.U. (C.O.) 2008]
w
1
Sol. (i) Probability of getting one head with one coin = .
2
FG 1IJ 6
1
∴ The probability of getting six heads with six coins =
H 2K =
64
1
∴ Average number of six heads with six coins in 6400 throws = np = 6400 × = 100
64
∴ The mean of the Poisson distribution = 100.
Approximate probability of getting six heads x times when the distribution is Poisson
λx e − λ (100) x . e −100
= = .
x! x!
www.cgaspirants.com
www.cgaspirants.com
(ii) Since 2 modes are given when λ is an integer, modes are λ – 1 and λ.
∴ λ–1=3 ⇒ λ=4
e −4 (4) 3
Probability (when r = 3) =
3!
e −4 (4) 4
Probability (when r = 4) =
4!
∴ Required Probability = P(r = 3 or 4) = P(r = 3) + P(r = 4)
e −4 (4) 3 e −4 (4) 4 64 –4
+= = e = 0.39073.
3! 4! 3
m
Example 13. For a Poisson distribution with mean m, show that
∞
dμ r e −m . m x
∑ ( x − m) r
o
μr+1 = mr μr–1 + m where, μr = . (U.P.T.U. 2007)
dm x!
x=0
.c
∞
e −m . m x
Sol. μr = ∑ ( x − m) r .
dμ r
=
x=0
∞
∑ x!
n x!
ts
LM − e −m . m x (x − m) r + e −m {xm x−1 (x − m) r − r(x − m) r−1 . m x }OP
x = 0 MN PQ
dm x!
ra
∞ ∞
dμ r e −m e −m x
⇒ m = ∑ mx (x – m)r+1 – rm ∑ m (x – m)r–1 = μr+1 – mr μr–1
pi
dm x! x!
x=0 x=0
dμ r
as
⇒ μr+1 = m+ mr μr–1.
dm
Example 14. Show that in a Poisson distribution with unit mean, mean deviation about
g
mean is
FG 2 IJ times the standard deviation. (G.B.T.U. 2012)
H eK
.c
Sol. Here, λ = 1
w
e −1 . (1) x e −1
w
∴ P(X = x) = = ; x = 0, 1, 2, ...
x! x!
Mean deviation about mean 1 is
w
∞ ∞
| x − 1| −1 1 2LM 3 OP
= ∑ | x − 1 | p( x) = e −1 ∑ x!
=e 1+ + +
N
2! 3! 4!
+ ...
Q ...(1)
x=0 x=0
n n +1−1 1 1
we have, = = −
(n + 1) ! (n + 1) ! n ! (n + 1) !
LM F 1 1 1IJ FG IJ OP
∴ From (1), Mean deviation about mean = e–1 1 + 1 −
MN GH 2!
+ −
2! 3!
+ ...
K H K PQ
2 2
= e–1 (1 + 1) = × 1 = × S.D. | Since Variance = mean = 1
3 e
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. If X is a Poisson variate such that P(X = 2) = 9P(X = 4) + 90P(X = 6), find the standard deviation.
2. If a random variable has a Poisson distribution such that P(1) = P(2), find
(i) mean of the distribution (ii) P(4).
2
3. Suppose that X has a Poisson distribution. If P(X = 2) = 3 P(X = 1) find, (i) P(X = 0) (ii) P(X = 3).
4. A certain screw making machine produces on average 2 defective screws out of 100, and packs
them in boxes of 500. Find the probability that a box contains 15 defective screws.
5. (i) The incidence of occupational disease in an industry is such that the workmen have a 10%
chance of suffering from it. What is the probability that in a group of 7, five or more will suffer
from it?
m
(ii) The experience shows that 4 industrial accidents occur in a plant on an average per month.
Calculate the probabilities of less than 3 accidents in a certain month. Use Poisson distribution.
(Given : e–4 = 0.01832). [M.T.U. (MBA) 2011]
o
6. (i) Suppose a book of 585 pages contains 43 typographical errors. If these errors are randomly
.c
distributed throughout the book, what is the probability that 10 pages, selected at random,
will be free from errors ?
ts
(ii) Assume that the probability of an individual coalminer being killed in a mine accident during
1
a year is . Use Poisson’s distribution to calculate the probability that in a mine employing
n
2400
200 miners there will be at least one fatal accident in a year.
ra
7. (i) A manufacturer of cotter pins knows that 5% of his product is defective. If he sells cotter pins
in boxes of 100 and guarantee that not more than 10 pins will be defective, what is the
approximate prob. that a box will fail to meet the guaranteed quality?
pi
(ii) An insurance company insures 4000 people against loss of both eyes in a car accident. Based
on previous data it was assumed 10 persons out of 1,00,000 will have such type of injury in car
accident. What is probability that more than 2 of the insured will collect on their policy in a
as
given year? (M.T.U. 2013)

8. Records show that the probability is 0.00002 that a car will have a flat tyre while driving over a
g
certain bridge. Use Poisson distribution to find the probability that among 20,000 cars driven
over the bridge, not more than one will have a flat tyre.
.c
9. Between the hours of 2 and 4 P.M., the average no. of phone calls per minute coming into the
switch board of a company is 2.5. Find the probability that during a particular minute, there will
w
be no phone call at all. [Given : e–2 = 0.13534 and e–0.5 = 0.60650.]

10. (i) Fit a Poisson distribution to the following data given the number of yeast cells per square for
w
400 squares :
No. of cells per sq. : 0 1 2 3 4 5 6 7 8 9 10
w
No. of squares : 103 143 98 42 8 4 2 0 0 0 0

It is given that e–1.3225 = 0.2665.
(ii) Data was collected over a period of 10 years, showing number of deaths from horse kicks in
each of the 200 army corps. The distribution of deaths was as follows:
No. of deaths : 0 1 2 3 4 Total
Frequency : 109 65 22 3 1 200
Fit a Poisson distribution to the data and calculate the theoretical frequencies.
[M.T.U. (B. Pharma) 2011 ; M.T.U. (MBA) 2011]
(iii) The following table gives the no. of days in a 50 day period during which automobile accidents
occured in a city.
No. of accidents : 0 1 2 3 4
No. of days : 21 18 7 3 1
Fit a Poisson distribution to the data. (G.B.T.U. 2011)
www.cgaspirants.com
www.cgaspirants.com
11. The number of accidents in a year involving taxi drivers in a city follows a Poisson distribution
with mean equal to 3. Out of 1000 taxi drivers, find approximately the number of drivers with
(i) no accidents (ii) more than 3 accidents in a year.
12. The distribution of the number of road accidents per day in a city is Poisson with mean 4. Find the
number of days out of 100 days when there will be
(i) no accident (ii) at least 2 accidents
(iii) atmost 3 accidents (iv) between 2 and 5 accidents.
13. It is given that 2% of the electric bulbs manufactured by a company are defective. Using Poisson
distribution, find the probability that a sample of 200 bulbs will contain (i) no defective bulb
(ii) two defective bulbs (iii) at the most three defective bulbs. (G.B.T.U. 2011)
14. A manager accepts the work submitted by his typist only when there is no mistake in the work.
m
The typist has to type on an average 20 letters per day of about 200 words each. Find the chance
of her making a mistake if
(i) less than 1% of the letters submitted by her are rejected
o
(ii) on 90% days all the letters submitted by her are accepted.
.c
As the probability of making a mistake is small, you may use Poisson distribution. (Take e = 2.72)
[Hint: (i) e–200p ≥ 0.99 (ii) e–4000p = 0.90]
15.
ts
In a certain factory manufacturing razor blades, there is a small chance of 0.002 for any blade to
be defective. The blades are supplied in packets of 10. Use suitable distribution to calculate the
n
approximate number of packets containing no defective, one defective and two defective blades
ra
respectively in a consignment of 20,000 packets. (M.T.U. 2014)
Answers
pi
2 32 –(4/3)
1. 1 2. (i) 2 (ii) 3. (i) e–(4/3) (ii)
e 4. 0.035
2 81
3e
as
5. (i) 0.0008 (ii) 0.23816 6. (i) 0.4795 (ii) 0.08 7. (i) 0.0136875 (ii) 0.007926
8. 0.938 9. 0.08208
10. (i) Theoretical frequencies are 107, 141, 93, 41, 13, 4, 1, 0, 0, 0, 0.
g
(0.61) r e −0.9 (0.9) r

.c
(ii) 108.67 × ; 109, 66, 20, 4 and 1 (iii) (50) ; 20, 18, 8, 3 and 1.
r! r!
w
11. (i) 50 (ii) 353

12. (i) 2 (ii) 91 (iii) 43 (iv) 39
w
13. (i) 0.018315 (ii) 0.146525 (iii) 0.43347

14. (i) p ≤ 0.0000506 (ii) p = 0.0000263 15. 19604, 392, 4
w
3.61 NORMAL DISTRIBUTION [U.P.T.U. 2007]
The normal distribution is a continuous distribution. It can be derived from the binomial distri-
bution in the limiting case when n, the number of trials is very large and p, the probability of a
success, is close to 12 . The general equation of the normal distribution is given by
FG
1 x−μ IJ 2
f(x) =
1
e
−
2H σ K
σ 2π
www.cgaspirants.com
www.cgaspirants.com
where the variable x can assume all values from – ∞ to + ∞. μ and σ, called the parameters of
the distribution, are respectively the mean and the standard deviation of the distribution and
– ∞ < μ < ∞, σ > 0. x is called the normal variate and f(x) is called probability density function of
the normal distribution.
If a variable x has the normal distribution with mean μ and standard deviation σ, we
briefly write x : N(μ, σ2).
The graph of the normal distribution is called the normal curve. It is bell-shapped and
symmetrical about the mean μ. The two tails of the curve extend to + ∞ and – ∞ towards the
positive and negative directions of the x-axis respectively and gradually approach the x-axis
without ever metting it. The curve is unimodal and f(x)
the mode of the normal distribution coincides with its
m
mean μ. The line x = μ divides the area under the
normal curve above x-axis into two equal parts. Thus,
o
the median of the distribution also coincides with its
mean and mode. The area under the normal curve
.c
between any two given ordinates x = x1 and x = x2
represents the probability of values falling into the given
interval. The total area under the normal curve above
the x-axis is 1.
n O m
ts x
ra
3.62 BASIC PROPERTIES OF THE NORMAL DISTRIBUTION
The probability density function of the normal distribution is given by

pi
FG
1 x−μ IJ 2
f(x) =
1
e
−
2 Hσ K
as
σ 2π
(i) f(x) ≥ 0 (ii) z ∞
f ( x) dx = 1 ,
g
–∞
i.e., the total area under the normal curve above the x-axis is 1.
.c
(iii) The normal distribution is symmetrical about its mean.

(iv) It is a unimodal distribution. The mean, mode and median of this distribution coincide.
w
3.63 STANDARD FORM OF THE NORMAL DISTRIBUTION

w
If X is a normal random variable with mean μ and f(z)

w
standard deviation σ, then the random variable Z =

X−μ
has the normal distribution with mean 0 and
σ
standard deviation 1. The random variable Z is called
the standardized (or standard) normal random vari-
able.
The probability density function for the nor-
O z
mal distribution in standard form is given by
(– ∞ < z < ∞)
1
1 − z2
f(z) = e 2
2π
www.cgaspirants.com
www.cgaspirants.com
It is free from any parameter. This helps us to compute areas under the normal probability
curve by making use of standard tables.
Note 1. If f(z) is the probability density function for the normal distribution, then
P(z1 ≤ Z ≤ z2) = zz1

z2
f(z) dz = F(z2) – F(z1), where F(z) = z
The function F(z) defined above is called the distribution function for the normal distribution.
−∞
z
f(z) dz = P(Z ≤ z)
Note 2. The probabilities P(z1 ≤ Z ≤ z2), P(z1 < Z ≤ z2), P(z1 ≤ Z < z2) and P(z1 < Z < z2) are all regarded to be
the same.
Note 3. F(– z1) = 1 – F(z1).
3.64 NORMAL DISTRIBUTION AS A LIMITING FORM OF BINOMIAL DISTRIBUTION
m
(when p = q) (U.P.T.U. 2007)
o
Let N (q + p)n be the binomial distribution. If p = q then p = q = (since p + q = 1) and
2
.c
consequently the binomial distribution is symmetrical. Let n be an even integer say 2k, k being
an integer. Since n → ∞, the frequencies of r and r + 1 successes can be written in following
forms:
n
f (r) = N. 2kCr
FG 1IJ
H 2K
2k ts
ra
2k FG 1IJ 2k
f (r + 1) = N . C r +1
H 2K
pi
2k
f (r + 1) C r +1 2k − r
∴ = =
as
2k
f (r) Cr r+1
The frequency of r successes will be greater than the frequency of (r + 1) successes if
g
f (r) > f (r + 1)
.c
f (r + 1)
⇒ <1
f ( r)
w
⇒ 2k – r < r + 1
w
1
⇒ r > k− ...(1)
2
w
In a similar way, the frequency of r successes will be greater than the frequencies of
1
(r – 1) successes if r < k + ...(2)
2
1 1
In view of (1) and (2), we observe that if k − < r < k + the frequency corresponding to
2 2
r successes will be the greatest. Clearly, r = k is the value of the success corresponding to which
the frequency is maximum. Suppose it is y0. Then, we have
2k FG 1IJ 2k
2k ! 1 FG IJ 2k
y0 = N . Ck
H 2K = N.
k! k! 2 H K
www.cgaspirants.com
www.cgaspirants.com
Let yx be the frequency of k + x successes then, we have
2k FG 1IJ 2k
FG 1IJ 2k
2k !
yx = N . C k+ x
H 2K = N.
H 2K .
(k + x) ! (k − x) !
yx k ! k! k (k − 1) (k − 2) ... (k − x + 1)
Now, = =
y0 ( k + x) ! (k − x) ! (k + x) (k + x − 1) ... (k + 1)
FG 1 − 1IJ FG 1 − 2 IJ ... RS1 − x − 1UV

=
H kK H kK T k W
FG 1 + 1IJ FG 1 + 2 IJ ... FG 1 + x IJ
H kK H kK H kK
m
Taking log on both sides,
LM FG IJ FG IJ
FG IJ OP
o
yx 1 2 x−1
log
N H
= log 1 − + log 1 −
K
+ ... + log 1 −
H KH KQ
.c
y0 k k k
L F 1I F 2 I F xIO
n N H k H ts
− Mlog G 1 + J + log G 1 + J + ... + log G 1 + J P
K k K H kK Q
x
...(3)
Now, writing expression for each term and neglecting higher powers of (very small
ra
k
quantity), we get from (3),
pi
yx 1 1
log
y0
=−
k
l
1 + 2 + 3 + ... + ( x − 1) −
k
q
1 + 2 + 3 + ... + ( x − 1) + x l q
as
2 x
=− {1 + 2 + 3 + ... + ( x − 1)} −
k k
g
2 ( x − 1) x x x2
.c
=− − =−
k 2 k k
w
2
∴ yx = y0 e − x /k
n k
w
2
⇒ yx = y0 e − x / 2σ 2 |∵ σ2 = npq = =
4 2
w
which is normal distribution.
3.65 MEAN AND VARIANCE OF NORMAL DISTRIBUTION (U.P.T.U. 2015)
1. The A.M. of a continuous distribution f(x) is given by
z∞
x f ( x) dx
z
−∞
A.M. ( x ) = ∞ | By definition
f ( x) dx
−∞
www.cgaspirants.com
www.cgaspirants.com
Consider the normal distribution with μ, σ as the parameters then
x = z−∞
∞
x.
σ 2π
1
e
−
2 σ
FG
1 x−μ
H
IJ
K
2
dx
Since z ∞
−∞
f ( x) dx
= area under normal curve = 1

x−μ
Put = z so that x = μ + σz ∴ dx = σdz
σ
so, x= z−∞
∞
(μ + σz)
σ 2π
1
e
1
− z2
2 (σ dz)
=μ z ∞ 1
e− z
2
/2
dz +
σ
z ∞
ze
1
− z2
2 dz
m
−∞ 2π 2π −∞
σ
z Fz I 2
z 1
o
∞ ∞
GH 2 JK
2 2
=μ+ e− z /2
d ∵ e− z /2
dz = 1
2π −∞ −∞ 2π
.c
Fe I − z2 / 2
∞
ts
σ
=μ+ G J
2π H − 1 K
−∞
n
x =μ
ra
2. By definition,
Variance =
z ∞
( x − x ) 2 f ( x) dx
pi
−∞
= z ∞
x 2 f ( x) dx + x 2 z ∞
f ( x) dx − 2 x z ∞
xf ( x) dx
as
−∞ −∞ −∞
z ∵ z ∞
f ( x) dx = 1 and
g
z
−∞
= x 2 f ( x) dx + x 2 − 2 x x ∞
.c
−∞ x f ( x) dx = x
−∞
z
w
∞
= x 2 f ( x) dx − x 2 ...(1)
−∞
w
z z
FG
1 x−x IJ 2
∞ ∞ 1 −
H K
w
2 2 2 σ
Now, Let I = x f ( x) dx = x e dx
−∞ −∞ σ 2π
x−x
Put = z so that x = x + σz ∴ dx = σ dz
σ
Hence, I= z−∞
∞
( x + σz) 2
σ 2π
1
e− z
2
/2
σ dz
=
1 LMσ
2π N
2
z ∞
−∞
z2 e− z
2
/2
dz + x 2 z ∞
−∞
e− z
2
/2
dz + 2σx z−∞
∞
z e− z
2
/2
dz
OP
Q
www.cgaspirants.com
www.cgaspirants.com
=
− σ2
2π z−∞
∞
z d( e − z
2
/2
) + x 2 . 1 + 2σx . 0
=−
σ2
2π
ez e − z2 / 2
2
j
∞
−∞
+
σ2
2π z
−∞
∞
e− z
2
/2
dz + x 2
=0+ σ2 .1+ x = σ2 + x2
∴ From (1), Variance = σ 2 + x 2 − x 2 = σ 2

∴ The standard deviation of the normal distribution is σ.
m
3.66 AREA UNDER THE NORMAL CURVE
x−μ
o
By taking z = , standard normal curve is formed. f(z)
σ
The total area under this curve is 1.
.c
The area under the curve is divided into two
equal parts by z = 0. The area between the ordinate
ts
Area = 0.5 Area = 0.5
z = 0 and any other ordinate can be noted from the
n
supplied table. It should be noted that mean for the
normal distribution is 0.
ra
O z
3.67 APPLICATIONS OF NORMAL DISTRIBUTION

pi
De Moivre made the discovery of this distribution in 1733.

as
This distribution has an important application in the theory of errors made by chance in
experimental measurements. Its more applications are in computation of hit probability of a
g
shot and in statistical inference in almost every branch of science.

.c
EXAMPLES
w
Example 1. A sample of 100 dry battery cells tested to find the length of life produced the
w
following results:
x = 12 hours, σ = 3 hours.
w
Assuming the data to be normally distributed, what percentage of battery cells are expected
to have life
(i) more than 15 hours (ii) less than 6 hours (iii) between 10 and 14 hours?
Sol. Here x denotes the length of life of dry battery cells.
x − x x − 12
Also z= = .
σ 3
(i) When x = 15, z = 1
∴ P(x > 15) = P(z > 1)
= P(0 < z < ∞) – P(0 < z < 1)
= .5 – 0.3413 = 0.1587 = 15.87%. 0 1
www.cgaspirants.com
www.cgaspirants.com
(ii) When x = 6, z = – 2
∴ P(x < 6) = P(z < – 2)
= P(z > 2) = P(0 < z < ∞) – P(0 < z < 2)
= 0.5 – 0.4772 = 0.0228 = 2.28%.
2
(iii) When x = 10, z = – = – 0.67 –2 0 2
3
2
When x = 14, z = = 0.67
3
P(10 < x < 14)
= P(– 0.67 < z < 0.67)
= 2P(0 < z < 0.67) = 2 × 0.2485
m
= 0.4970 = 49.70%. –.67 0 .67
Example 2. In a sample of 1000 cases, the mean of a certain test is 14 and S.D. is 2.5.
o
Assuming the distribution to be normal, find
(i) how many students score between 12 and 15?
.c
(ii) how many score above 18?
(iii) how many score below 8?
(iv) how many score 16?
Sol. (i)
n
x1 − μ 12 − 14
z1 = = = – 0.8
ts
ra
σ 2.5
x2 − μ 15 − 14
z2 = = = 0.4
σ 2.5
pi
Area lying between – 0.8 and 0.4

= Area between 0 to 0.8 + Area between 0 to 0.4
as
= 0.2881 + 0.1554 = 0.4435

Reqd. no. of students = 1000 × 0.4435 = 444 (app.)
g
18 − 14
(ii) z= = 1.6
.c
2.5
Area right to 1.6 = 0.5 – (Area between 0 and 1.6) = 0.5 – 0.4452 = 0.0548
w
Reqd. no. of students = 1000 × 0.0548 = 54.8 ≈ 55 (app.)

w
8 − 14
(iii) z= = – 2.4
2.5
w
Area left to – 2.4 = 0.5 – (Area between 0 and 2.4) = 0.5 – 0.4918 = 0.0082
∴ Reqd. no. of students = 1000 × 0.0082 = 8.2 ≈ 8 (app.).
15.5 − 14
(iv) z1 = = 0.6
2.5
16.5 − 14
z2 = =1
2.5
Area between 0.6 and 1 = 0.3413 – 0.2257 = 0.1156
∴ Reqd. no. of students = 1000 × 0.1156 = 115.6 ≈ 116 (app.).
www.cgaspirants.com
www.cgaspirants.com
Example 3. Assume mean height of soldiers to be 68.22 inches with a variance of 10.8
inches square. How many soldiers in a regiment of 1,000 would you expect to be over 6 feet tall,
given that the area under the standard normal curve between z = 0 and z = 0.35 is 0.1368 and
between z = 0 and z = 1.15 is 0.3746. [G.B.T.U. (C.O.) 2011]
Sol. x = 6 feets = 72 inches
x − μ 72 − 68.22
∴ z= = = 1.15
σ 10.8
P(x > 72) = P(z > 1.15) = 0.5 – P(0 ≤ z ≤ 1.15)
= 0.5 – 0.3746 = 0.1254
∴ Expected no. of soldiers = 1000 × 0.1254 = 125.4 ≈ 125 (app.).
m
Example 4. A large number of measurement is normally distributed with a mean 65.5″
and S.D. of 6.2″. Find the percentage of measurements that fall between 54.8″ and 68.8″.
Sol. Mean μ = 65.5 inches, S.D. σ = 6.2 inches
o
f(z)
x1 = 54.8 inches, x2 = 68.8 inches
.c
x1 − μ 54.8 − 65.5
∴ z1 = = = – 1.73
σ 6.2
and
Now,
z2 =
x2 − μ 68.8 − 65.5
σ
=
6.2
n = 0.53 ts – 1.73
P(– 1.73 ≤ z ≤ 0.53) = P(– 1.73 ≤ z ≤ 0) + P(0 ≤ z ≤ 0.53)
0 .53 z
ra
= P(0 ≤ z ≤ 1.73) + P(0 ≤ z ≤ 0.53)

= 0.4582 + 0.2019 = 0.6601 | By table
pi
∴ Reqd. percentage of measurements = 66.01%.

Example 5. A manufacturer knows from experience that the resistance of resistors he
produces is normal with mean μ = 100 ohms and standard deviation σ = 2 ohms. What percentage
as
of resistors will have resistance between 98 ohms and 102 ohms?

Sol. μ = 100 Ω, σ = 2 Ω, x1 = 98 Ω, x2 = 102 Ω
g
x1 − μ 98 − 100
.c
∴ z1 = = =–1
σ 2 f(z)
w
x2 − μ 102 − 100
and z2 = = = 1.
σ 2
w
Now, P(98 < x < 102) = P(– 1 < z < 1)

= P(– 1 ≤ z ≤ 0) + P(0 ≤ z ≤ 1)
w
= P(0 ≤ z ≤ 1) + P(0 ≤ z ≤ 1)
= 0.3413 + 0.3413 = 0.6826. –1 O 1 z
∴ Percentage of resistors having resistance be-
tween 98 Ω and 102 Ω = 68.26%.
Example 6. In a normal distribution, 31% of the items are under 45 and 8% are over 64.
Find the mean and standard deviation of the distribution. It is given that if f(t) =
then f(0.5) = 0.19 and f(1.4) = 0.42.

1
2π 0z t
e
dx
(M.T.U. 2013)
−
1 2
2
x
Sol. Let μ and σ be the mean and S.D. respectively.

31% of the items are under 45.
www.cgaspirants.com
www.cgaspirants.com
⇒ Area to the left of the ordinate x = 45 is 0.31

When x = 45, let z = z1
P(z1 < z < 0) = 0.5 – 0.31 = 0.19
From the tables, the value of z corresponding to this
area is 0.5
∴ z1 = – 0.5[z1 < 0] 31% 8%
When x = 64, let z = z2
P(0 < z < z2) = 0.5 – 0.08 = 0.42 x = 45 x=μ x = 64
z = z1 z=0 z = z2
From the tables, the value of z corresponding to this
area is 1.4.
m
∴ z2 = 1.4
x−μ
o
Since z=
σ
.c
45 − μ 64 − μ
∴ – 0.5 = and 1.4 =
σ σ
and
⇒
Subtracting
45 – μ = – 0.5σ
64 – μ = 1.4σ
n
– 19 = – 1.9σ ∴ σ = 10
ts ...(1)
...(2)
ra
From (1), 45 – μ = – 0.5 × 10 = – 5 ∴ μ = 50.

Example 7. The life of army shoes is normally distributed with mean 8 months and
pi
standard deviation 2 months. If 5000 pairs are insured, how many pairs would be expected to
LM
need replacement after 12 months? Given that P(z ≥ 2) = 0.0228 and z =
x−μ
.
OP
N Q
as
σ
Sol. Mean (μ) = 8, Standard Deviation (σ) = 2
Number of pairs of shoes = 5000, Total months (x) = 12
g
x − μ 12 − 8
.c
when x = 12, z= = =2
σ 2
w
Area (z ≥ 2) = 0.0228
Number of pairs whose life is more than 12 months = 5000 × 0.0228 = 114
w
Pair of shoes needing replacement after 12 months = 5000 – 114 = 4886.

Example 8. The mean inside diameter of a sample of 200 washers produced by a machine
w
is 0.502 cm and the standard deviation is 0.005 cm. The purpose for which these washers are
intended allows a minimum tolerance in the diameter of 0.496 to 0.508 cm, otherwise the washers
are considered defective. Determine the percentage of defective washers produced by the machine.
Assume the diameters are normally distributed.
Sol. Given: Mean μ = 0.502 cm, S.D. σ = 0.005 cm, x1 = 0.496 cm, x2 = 0.508 cm.
x1 − μ 0.496 − 0.502
Now, z1 = = = – 1.2
σ 0.005
x2 − μ 0.508 − 0.502
z2 = = = 1.2
σ 0.005
www.cgaspirants.com
www.cgaspirants.com
Area for non-defective washers

= P(– 1.2 ≤ z ≤ 1.2) f(z)
= P(– 1.2 ≤ z ≤ 0) + P(0 ≤ z ≤ 1.2)
= P(0 ≤ z ≤ 1.2) + P(0 ≤ z ≤ 1.2)
= 0.3849 + 0.3849 = 0.7698
= 76.98%.
∴ Percentage of defective washers = 100 – 76.98
– 1.2 O 1.2
= 23.02%. z
Example 9. Assuming that the diameters of 1000 brass plugs taken consecutively from a
machine, form a normal distribution with mean 0.7515 cm and standard deviation 0.002 cm,
how many of the plugs are likely to be rejected if the approved diameter is 0.752 ± 0.004 cm.
m
Sol. Tolerance limits of the diameter of non-defective plugs are
0.752 – 0.004 = 0.748 cm. and 0.752 + 0.004 = 0.756 cm.
o
x−μ
.c
Standard normal variable, z=
σ
If x1 = 0.748, z1 =
0.748 − 0.7515
0.002
0.756 − 0.7515
n = – 1.75
ts
If x2 = 0.756, z2 = = 2.25
ra
0.02
Area from (z1 = – 1.75) to (z2 = 2.25)
pi
= P(– 1.75 ≤ z ≤ 2.25) = P(– 1.75 ≤ z ≤ 0) + P(0 ≤ z ≤ 2.25)

= P(0 ≤ z ≤ 1.75) + P(0 ≤ z ≤ 2.25) = 0.4599 + 0.4878 = 0.9477
as
Number of plugs which are likely to be rejected = 1000 × (1 – 0.9477) = 1000 × .0523 = 52.3
Hence approximately 52 plugs are likely to be rejected.
g
Example 10. If the heights of 300 students are normally distributed with mean 64.5 inches
and standard deviation 3.3 inches, find the height below which 99% of the students lie.
.c
Sol. Mean μ = 64.5 inches, S.D. σ = 3.3 inches

w
x − 64.5
Area between 0 and = 0.99 – 0.5 = 0.49
3.3
w
From the table, for the area 0.49, z = 2.327

w
The corresponding value of x is given by

x − 64.5
= 2.327
3.3
⇒ x – 64.5 = 7.68
⇒ x = 7.68 + 64.5 = 72.18 inches.
Hence 99% students are of height less than 6 ft. 0.18 inches.
Example 11. The income of a group of 10,000 persons was found to be normally distributed
with mean ` 750 p.m. and standard deviation of ` 50. Show that, of this group, about 95% had
income exceeding ` 668 and only 5% had income exceeding ` 832. Also find the lowest income
among the richest 100.
Sol. Given: μ = 750, σ = 50
www.cgaspirants.com
www.cgaspirants.com
x−μ
Standard normal variable, z =
σ
x1 − μ 668 − 750
(i) If x1 = 668, z1 = = = – 1.64
σ 50
P(x1 > 668) = P(z1 > – 1.64)
= 0.5 + P(– 1.64 ≤ z ≤ 0) f(z)
= 0.5 + P(0 ≤ z ≤ 1.64)

= 0.5 + 0.4495
= 0.9495
∴ Required percentage of persons having income
m
exceeding ` 668 = 94.95% ≈ 95% (approx.) z1 = – 1.64 O z
x2 − μ 832 − 750
o
(ii) If x2 = 832, z2 = = = 1.64 f(z)
σ 50
.c
P(x2 > 832) = P(z2 > 1.64)
= 0.5 – P(0 ≤ z ≤ 1.64)
= 0.5 – 0.4495 = 0.0505
∴ Required percentage of persons having in-
come exceeding ` 832 = 5.05% ≈ 5% (approx.)
n ts O z2 = 1.64 z
ra
(iii) Let x be the lowest income among the richest 100 persons i.e., 1% of 10,000.
Thus, area between O and z = 0.49 (see figure) by Normal distribution table,
pi
z = 2.33 f(z)
x−μ
Thus, = 2.33
as
σ
0.01
x − 750
⇒ = 2.33
g
50
⇒ x = 866.5 O z z
.c
Hence ` 866.5 is the minimum income among the richest 100 persons.
Example 12. 255 metal rods were cut roughly 6 inches over size. Finally the lengths of
w
the over size amount, were measured exactly and grouped with 1 inch intervals, there being in
w
1″ 1″ 1″ 1″ 1″ 1″
all 12 groups –1 ,1 – 2 , ...... , 11 – 12 .
2 2 2 2 2 2
w
The frequency distribution for the 255 lengths was as follows:
Length (inches) 1 2 3 4 5 6 7 8 9 10 11 12
Central value
Frequency 2 10 19 25 40 44 41 28 25 15 5 1
Fit a normal curve to this data.

Sol. The equation of the normal curve for N observations is
FG
1 x−μ IJ 2
y=
N
e
−
2H σ K ...(1)
σ 2π
www.cgaspirants.com
www.cgaspirants.com
x f u=x–6 fu fu2
1 2 –5 – 10 50
2 10 –4 – 40 160
3 19 –3 – 57 171
4 25 –2 – 50 100
5 40 –1 – 40 40
6 44 0 0 0
7 41 1 41 41
8 28 2 56 112
9 25 3 75 225
10 15 4 60 240
m
11 5 5 25 125
12 1 6 6 36
o
Total 255 66 1300
.c
Σfu 66
Mean, μ=a+ n
Σfu2
Σf
Σfu
=6+
255
= 6.259
FG IJ 2
1300 66
ts FG IJ 2
σ2 =
Variance, −
H K = −
H K = 5.031
ra
Σf Σf 225 255
∴ σ = 2.243
pi
Thus, we have N = 255, Mean, μ = 6.259″, S.D. σ = 2.243″

Hence the fitted curve is
as
FG
1 x − 6.259 IJ 2
y=
255
e
−
2 H
2.243 K | From (1)
g
2.243 2π
113.68
.c
2
= e − 0.099( x − 6.259) .
2π
w
Example 13. Show that the area under the normal curve is unity.
Sol. Area under the normal curve is given by
w
z
( x − μ)2
∞ 1 −
2σ 2
w
A= e dx
−∞ σ 2π
x−μ
Put = z so dx = σ dz
σ
∴ A= z −∞
∞
σ 2π
1
e− z
2
/2
(σdz) =
2
π z
0
∞
e− z
2
/2
dz
Now, A . A = A2 =
F
GH 2
π z0
∞
e− x
2
/2
IF
JK GH
dx
2
π z∞
0
e− y
2
/2
I
JK
dy
Fx I
zz
2
+ y2
2 ∞ ∞ − GH 2 JK
= e dx dy | where x and y are dummy variables
π 0 0
www.cgaspirants.com
www.cgaspirants.com
Put x = r cos θ, y = r sin θ so that J = r changing to polar coordinates,
A2 =
2
π zz
0
π/2 ∞
0
e− r
2
/2
r dr dθ = z
0
∞
e−r
2
/2
d
Fr I =1
2
GH 2 JK
∴ A = Area under the normal curve = 1
Example 14. Prove that for normal distribution, the mean deviation from the mean
4
equals to of the standard deviation approximately. (U.P.T.U. 2009)
5
Sol. Let μ and σ be the mean and standard deviation of the normal distribution. Then by
definition,
Mean deviation from the mean
m
=
z ∞
|x − μ| f ( x) dx
o
−∞
z
.c
( x − μ)2
1 ∞ −
2σ 2
= |x − μ| e dx
σ 2π −∞
=
σ 2π
1
z ∞
−∞
n
σ|z| e
−
1 2
2
z
σ dz
ts when
x−μ
σ
=z
⇒ dx = σ dz
ra
=σ
2
π z∞
0
z e− z
2
/2
dz
pi
2 2 ∞ 2 4
=σ − e− z / 2 = σ = 0.7979 σ ≈ 0.8σ ≈ σ
π π 5
as
ASSIGNMENT
g
.c
1. In a test on 2000 electric bulbs, it was found that the life of a particular make, was normally
distributed with an average life of 2040 hours and S.D. of 60 hours, estimate the number of bulbs
w
likely to burn for

(i) more than 2150 hours (ii) less than 1950 hours
w
(iii) more than 1920 hours but less than 2160 hours. (U.P.T.U. 2008)
2. An aptitude test for selecting officers in a bank is conducted on 1000 candidates. The average
w
score is 42 and the standard deviation of score is 24. Assuming normal distribution for the scores,
find
(i) the number of candidates whose scores exceed 60
(ii) the number of candidates whose scores lie between 30 and 60.
3. (i) In a normal distribution exactly normal, 7% of the items are under 35 and 89% are under 63.
What are the mean and standard deviation of the distribution? (G.B.T.U. 2010)
(ii) In a normal distribution, 0.0107 of the items lie below 42 and 0.0446 of the items lie above 82.
What is the mean and standard deviation of the normal distribution?
[U.P.T.U. (MBA) 2009]
4. If Z is a standard normal variable, find the following probabilities: [G.B.T.U. (MBA) 2010]
(i) P(Z < 1.2) (ii) P(Z > – 1.2) (iii) P(– 1.2 < Z < 1.3).
www.cgaspirants.com
www.cgaspirants.com
5. An aptitude test was conducted on 900 employees of the Metro Tyres Limited in which the mean
score was found to be 50 units and standard deviation was 20. On the basis of this information,
you are required to answer the following questions:
(i) What was the number of employees whose mean score was less than 30?
(ii) What was the number of employees whose mean score exceeded 70?
(iii) What was the number of employees whose mean score were between 30 and 70?
x−μ
0.25 0.50 0.70 1.00 1.25 1.50
σ
Area 0.0987 0.1915 0.2734 0.3413 0.3944 0.4332 [U.P.T.U. (MBA) 2009]
6. (a) Students of a class were given a mechanical aptitude test. Their marks were found to be
normally distributed with mean 60 and standard deviation 5. What percent of students
scored?
m
(i) more than 60 marks? (ii) less than 56 marks? (iii) between 45 and 65 marks?
(b) 2000 students appeared in an examination. Distribution of marks is assumed to be normal
o
with mean μ = 30 and σ = 6.25. How many students are expected to get marks?
.c
(i) between 20 and 40 (ii) less than 35 and (iii) above 50.
[U.P.T.U. (MBA) 2012]
7.
(i) between 69 and 74 kg
ts
(c) Suppose the weight W of 600 male students are normally distributed with mean μ = 70 kg and
standard deviation σ = 5 kg. Find number of students with weight
n
(ii) more than 76 kg. (G.B.T.U. 2013)
(a) In an intelligence test administered to 1000 students, the average score was 42 and standard
ra
deviation 24. Find:
(i) the expected number of students scoring more than 50.
(ii) the number of students scoring between 30 and 54.
pi
(iii) the value of score exceeded by top 100 students. [G.B.T.U. (MBA) 2010]
(b) The average monthly sales of 5000 firms are normally distributed. Its mean and standard
as
deviation are ` 36000 and ` 10000 respectively. Find:

(i) the no. of firms having sales over ` 40000.
g
(ii) the no. of firms having sales between ` 30000 and ` 40000.
[Given area under normal curve from 0 to z for Z (0.4) = 0.1554 and Z (0.6) = 0.2257]
.c
[G.B.T.U. (MBA) 2010]

(c) The daily wages of 1000 workers are distributed around a mean of ` 140 and with a standard
w
deviation of ` 10. Estimate the number of workers whose daily wages will be
(i) between ` 140 and ` 144 (ii) less than ` 126
w
(iii) more than ` 160. (G.B.T.U. 2012)

8. (a) Records kept by the goods inwards department of a large factory show that the average no. of
w
lorries arriving each week is 248. It is known that the distribution approximates to be normal
with a standard deviation of 26.
If this pattern of arrival continues, what percentage of weeks can be expected to have number
of arrivals of:
(i) less than 229 per week? (ii) more than 280 per week?
(b) Pipes for tobacco are being packed in fancy plastic boxes. The length of the pipe is normally
distributed with μ = 5″ and σ = 0.1″. The internal length of the boxes is 5.2″. What is the
probability that the box would be small for the pipe?
[Given that : φ(1.8) = 0.9641, φ(2) = 0.9772, φ(2.5) = 0.9938]
(c) A manufacturer of envelopes knows that the weight of the envelopes is normally distributed
with mean 1.9 gm and variance 0.01 square gm. Find how many envelopes weighing
(i) 2 gm or more
www.cgaspirants.com
www.cgaspirants.com
(ii) 2.1 gm or more, can be expected in a given packet of 1000 envelopes? (U.P.T.U. 2015)
[Given: if t is the normal variable, then φ(0 ≤ t ≤ 1) = 0.3413 and φ (0 ≤ t ≤ 2) = 0.4772]
9. (a) East-East Airlines has the policy of employing only Indian woman whose height is between
62 inches and 69 inches. If the height of Indian women is approximately normally distributed
with a mean of 64 inches and a standard deviation of 3 inches, Out of the 1000 applications
received find the number of applicants that would be
(i) too tall (ii) too short (iii) of acceptable height.
(b) The mean height of 500 students is 151 cm and the standard deviation is 15 cm. Assuming
that the heights are normally distributed, find how many students have heights between 120
and 155 cm?
(c) The monthly mess bill of a student who is staying in the hostel follows a normal distribution
with a mean of ` 2000 and a standard deviation of ` 185. What is the probability that in the
m
next month, his bill will go above ` 2400? [U.P.T.U. (MBA) 2009]
10. In an examination taken by 500 candidates, the average and the standard deviation of marks
obtained (normally distributed) are 40% and 10%. Find approximately
o
(i) how many will pass, if 50% is fixed as a minimum?
.c
(ii) what should be the minimum if 350 candidates are to pass?
(iii) how many have scored marks above 60%?
many workers were there in all?

n ts
11. The income distribution of workers in a certain factory was found to be normal with mean
` 500 and standard deviation equal to ` 50. There were 228 persons getting above ` 600. How
(Area under the standard normal curve between height at 0 and 2 is 0.4772).
ra
12. In a certain examination, the percentage of passes and distinction were 46 and 9 respectively.
Estimate the average marks obtained by the candidates, the minimum pass and distinction
marks being 40 and 75 respectively. Assume the distribution of marks to be normal.
pi
LMHint. 40 − μ = 0.1 and

75 − μ
= 1.34.
OP
N σ σ Q
as
13. The marks obtained by a no. of students for a certain subject are assumed to be approximately
normally distributed with mean value 65 and S.D. of 5. If 3 students are taken at random from
this set, what is the probability that exactly 2 of them will have marks over 70?
g
14. In an examination, it is laid down that a student passes if he secures 30% or more marks. He is
.c
placed in the first, second or third division according as he secures 60% or more marks, between
45% and 60% marks and marks between 30% and 45% respectively. He gets distinction incase he
secures 80% or more marks. It is noticed from the result that 10% of the students failed in the
w
examination, whereas 5% of them obtained distinction.

Calculate the percentage of students placed in the second division. (Assume normal distribution
w
of marks).
[Hint. P( X < 30) = 0.10, P(X ≥ 80) = 0.05]
w
15. Determine the minimum marks a student must get in order to receive an A grade if the top 10%
of the students are awarded A grades in an examination where the mean mark is 72 and standard
deviation is 9.
16. (i) When the mean of marks was 50% and S.D. 5% then 60% of the students failed in an
examination. Determine the ‘grace’ marks to be awarded in order to show that 70% of the
students passed. Assume that the marks are normally distributed.
LMHint. x − 0.5 = 0.25,
1 x2 − 0.5
= − 0.52
OP
N 0.05 0.05 Q
(ii) In a university examination of a particular year, 60% of the students failed when mean of the
marks was 50% and S.D. 5%. University decided to relax the conditions of passing by lowering
the pass marks to show its result 70%. Find the minimum marks for a student to pass
supposing the marks to be normally distributed and no change in the performance of students
takes place.
www.cgaspirants.com
www.cgaspirants.com
17. How does a normal distribution differ from a binomial distribution? What are the important
properties of normal distribution? [M.T.U. (MBA) 2012]
18. If the skulls are classified as A, B and C according as the length-breadth index is under 75,
between 75 and 80 or over 80, find approximately (assuming that the distribution is normal) the
mean and standard deviation of a series in which A are 58%, B are 38% and C are 4%, being given
that if f(t) =
1
2π z t
0
e− ( x
2
[Hint: P(X < 75) = 0.58, P(X > 80) = 0.04]

/2)
dx then f(0.20) = 0.08 and f(1.75) = 0.46.
19. The following table gives frequencies of occurrence of a variable X between certain limits:
Variable X Frequency
Less than 40 30
40 or more but less than 50 33
m
50 and more 37
The distribution is exactly normal. Find the distribution and also obtain the fequency between
o
X = 50 and X = 60.
.c
20. The marks X obtained in Mathematics by 1000 students are normally distributed with mean
78% and standard deviation 11%.
Determine:
(i) how many students got marks above 90%?
n
(ii) What was the highest marks obtained by the lowest 10% of students?
(iii) Within what limits did the middle 90% of the students lie?
ts
ra
Answers
1. (i) 67 (ii) 134 (iii) 1909 2. (i) 227 (ii) 465
pi
3. (i) x = 50.3, σ = 10.33 (ii) μ = 65, σ = 10

4. (i) 0.8849 (ii) 0.8849 (iii) 0.7881
as
5. (i) 143 (ii) 143 (iii) 614

6. (a) (i) 50%, (ii) 21.2%, (iii) 84% (b) (i) 1781, (ii) 1576, (iii) 1 (c) (i) 220, (ii) 69
7. (a) (i) 371, (ii) 383, (iii) 72.72 (b) (i) 1723 (ii) 1906 (c) (i) 155, (ii) 81, (iii) 23
g
8. (a) (i) 23% (ii) 11% (b) 0.0228 (c) (i) 159 (ii) 23
.c
9. (a) (i) 48 (ii) 251 (iii) 701 (b) 294 (c) 0.0154
10. (i) 79 (ii) 35% (iii) 11 11. 10,000 12. 37.2
w
13. 0.06357 14. 34% 15. 84 marks

16. (i) 3.85 (ii) 47.4. 18. μ = 74.35, σ = 3.23 19. μ = 46.12, σ = 11.76, 25
w
20. (i) 138 (ii) 63.92% (iii) between 60 and 96.

w
3.68 POPULATION OR UNIVERSE
An aggregate of objects (animate or inanimate) under study is called population or universe.

It is thus a collection of individuals or of their attributes (qualities) or of results of operations
which can be numerically specified.
A universe containing a finite number of individuals or members is called a finite inverse.
For example, the universe of the weights of students in a particular class.
A universe with infinite number of members is known as an infinite universe. For
example, the universe of pressures at various points in the atmosphere.
In some cases, we may be even ignorant whether or not a particular universe is infinite,
e.g., the universe of stars.
www.cgaspirants.com
www.cgaspirants.com
The universe of concrete objects is an existent universe. The collection of all possible
ways in which a specified event can happen is called a hypothetical universe. The universe
of heads and tails obtained by tossing a coin an infinite number of times (provided that it does
not wear out) is a hypothetical one.
3.69 SAMPLING
The statistician is often confronted with the problem of discussing universe of which he cannot
examine every member i.e., of which complete enumeration is impracticable. For example, if
we want to have an idea of the average per capita income of the people of India, enumeration of
every earning individual in the country is a very difficult task. Naturally, the question arises :
What can be said about a universe of which we can examine only a limited number of members
m
? This question is the origin of the Theory of Sampling.
A finite subset of a universe is called a sample. A sample is thus a small portion of the
universe. The number of individuals in a sample is called the sample size. The process of
o
selecting a sample from a universe is called sampling.
.c
The theory of sampling is a study of relationship existing between a population and
samples drawn from the population. The fundamental object of sampling is to get as much
ts
information as possible of the whole universe by examining only a part of it. An attempt is thus
made through sampling to give the maximum information about the parent universe with the
minimum effort.
n
ra
Sampling is quite often used in our day-to-day practical life. For example, in a shop we
assess the quality of sugar, rice or any other commodity by taking only a handful of it from the
bag and then decide whether to purchase it or not. A housewife normally tests the cooked
pi
products to find if they are properly cooked and contain the proper quantity of salt or sugar, by
taking a spoonful of it.
as
3.70 SAMPLING METHODOLOGIES

g
Sampling methodologies are classified under two general categories:

1. Probability sampling and 2. Non-probability sampling
.c
In the former, the researcher knows the exact possibility of selecting each member of
w
the population while in the latter, the chance of being included in the sample is not known. A
probability sample tends to be more difficult and costly to conduct. However, probability samples
w
are the only type of samples where the results can be generalized from the sample to the
population. In addition, probability samples allow the researcher to calculate the precision of
w
the estimates obtained from the sample and to specify the sampling error.
Non-probability samples, incontrast, donot allow the study’s findings to be generalized
from the sample to the population. When discussing the results of a non-probability sample,
the researchers must limit his/her findings to the persons or elements sampled.
This procedure also does not allow the researcher to calculate sampling statistics that
provide information about the precision of the results. The advantage of non-probability sam-
pling is the case in which it can be administered.
Non-probability samples tend to be less complicated and less time consuming than prob-
ability samples. If the researcher has no intention of generalizing beyond the sample, one of
the non-probability sampling methodologies will provide the desired information.
www.cgaspirants.com
www.cgaspirants.com
3.71 NON-PROBABILITY SAMPLING
The three common types of non-probability samples are:

(i) Convenience Sampling. As the name implies, convenience sampling involves choosing
respondents at the convenience of the researcher. Examples of convenience sampling include
people-in-the street interviews—the sampling of people to which the researcher has easy access,
such as a class of students and studies that use people who have volunteered to be questioned
as a result of an advertisement or another type of promotion. A drawback to this methodology
is the lack of sampling accuracy. Because the probability of inclusion in the sample is unknown
for each respondent, none of the reliability or sampling precision statistics can be calculated.
Convenience samples, however, are employed by researchers because the time and cost of
collecting information can be reduced.
m
(ii) Quota Sampling [G.B.T.U. (B. Pharm.) 2010]
Quota sampling is often confused with stratified and cluster sampling—two probability
o
sampling methodologies. All of these methodologies sample a population that has been
subdivided into classes or categories.
.c
The primary differences between the methodologies is that with stratified and cluster
ts
sampling, the classes are mutually exclusive and are isolated prior to sampling. Thus, the
probability of being selected is known and members of the population selected to be sampled
are not arbitrarily disqualified from being included in the results. In quota sampling, the
n
classes cannot be isolated prior to sampling and respondents are categorized into the classes
ra
as the survey proceeds. As each class fills or reaches its quota, additional respondents that
would have fallen into these classes are rejected or excluded from the results.
An example of a quota sample would be a survey in which the researcher desires to
pi
obtain a certain number of respondents from various income categories. Generally, researchers
donot know the income of the persons they are sampling until they ask about income. Therefore,
as
the researcher is unable to subdivide the population from which the sample is drawn into
mutually exclusive income categories prior to drawing the sample.
(iii) Judgemental Sampling. In judgemental or purposive sampling, the researcher em-
g
ploys his or her own expert judgement about who to include in the sample frame. Prior knowl-
.c
edge and research skill are used in selecting the respondents or elements to be sampled.
An example of this type of sample would be a study of potential users of a new recreational
w
facility that is limited to those persons who live within two miles of the new facility. Expert
judgement based on past experience indicates that most of the use of this type of facility comes
w
from persons living within two miles. However, by limiting the sample to only this group,
usage projections may not be reliable if the usage characteristics of the new facility vary from
w
those previously experienced. As with all non-probability sampling methods, the degree and
direction of error introduced by the researcher cannot be measured and statistics that measure
the precision of the estimates cannot be calculated.
3.72 PROBABILITY SAMPLING
Five methodologies are most commonly used for conducting probability sampling.
(i) Simple Random Sampling. Simple random sampling provides the base from which the
other more complex sampling methodologies are derived.
To conduct a simple random sampling, the researcher must first prepare an exhaustive
list (sampling frame) of all members of the population of interest. From this list, the sample is
drawn so that each person or item has an equal chance of being drawn during each selection
www.cgaspirants.com
www.cgaspirants.com
round. Samples may be drawn with or without replacement. In practice, however, most simple
random sampling for survey research is done without replacement ; that is, a person or item
selected for sampling is removed from the population for all subsequent selections. At any
draw, the process for a simple random sample without replacement must provide an equal
chance of inclusion to any member of the population not already drawn. To draw a simple
random sample without introducing researcher bias, computerized sampling programs and
random numbers tables are used to impartially select the members of the population to be
sampled.
An example of a simple random sample would be a survey of County employees. An
exhaustive list of all County employees as of a certain date could be obtained from the
Department of Human Resources. If 100 names were selected from this list using a random
number table or a computerized sampling program, then a simple random sample would be
m
created. Such a random sampling procedure has the advantage of reducing bias and enables
the researcher to estimate sampling errors and the precision of the estimates derived through
o
statistical calculations.
(ii) Stratified Random Sampling [G.B.T.U. (B. Pharm.) 2010]
.c
Stratified random sampling involves categorizing the members of the population into mutually
exclusive and collectively exhaustive groups. An independent simple random sample is then
ts
drawn from each group. Stratified sampling techniques can provide more precise estimates if
the population being surveyed is more heterogeneous than the categorized groups, can enable
n
the researcher to determine desired levels of sampling precision for each group, and can provide
ra
administrative efficiency.
An example of a stratified sample would be a sample conducted to determine the average
income earned by families in the United States. To obtain more precise estimates of income,
pi
the researcher may want to stratify the sample by geographic region (northeast, mid-Atlantic,
etc.) and/or stratify the sample by urban, suburban, and rural groupings. If the differences in
as
income among the regions or groupings are greater than the income differences within the
regions or groupings, precision of the estimates is improved. In addition, if the resarch
organization has branch offices located in these regions, the administration of the survey can
g
be decentralized and perhaps conducted in a more cost-efficient manner.

.c
(iii) Cluster Sampling. Cluster sampling is similar to stratified sampling because the
population to be sampled is subdivided into mutually exclusive groups. However, in cluster
w
sampling, the groups are defined so as to maintain the heterogeneity of the population. It is
the researcher’s goal to establish clusters that are representative of the population as a whole,
w
although in practice this may be difficult to achieve. After the clusters are established, a
simple random sample of the clusters is drawn and the members of the chosen clusters are
w
sampled. If all of the elements (members) of the clusters selected are sampled, then the sampling
procedure is defined as one-stage cluster sampling. If a random sample of the elements of
each selected cluster is drawn, then the sampling procedure is defined as two-stage cluster
sampling.
Cluster sampling is frequently employed when the researcher is unable to compile a
comprehensive list of all the elements in the population of interest. A cluster sample might be
used by a researcher attempting to measure the age distribution of persons residing in Mumbai.
It would be much more difficult for the researcher to compile a list of every person residing in
Mumbai than to compile a list of residential addresses. In this example, each address would
represent a cluster of elements (persons) to be sampled. If the elements contained in the clusters
are as heterogeneous as the population, then estimates derived from cluster sampling are as
precise as those from simple random sampling. However, if the heterogeneity of the clusters is
less than that of the population, the estimates will be less precise.
www.cgaspirants.com
www.cgaspirants.com
(iv) Systematic Sampling. Systematic sampling, a form of one-stage cluster sampling, is

often used in place of simple random sampling. In systematic sampling, the researcher selects
every nth member after randomly selecting the first through nth element as the starting point.
For example, if the researcher decides to sample every 20th member of the population, a 5
percent sample, the starting point for the sample is randomly selected from the first 20 members.
A systematic sample is a type of cluster sample because each of the first 20 members of the
sampling frame defines a cluster that contains 5 percent of the population.
A researcher may choose to conduct a systematic sample instead of a simple random
sample for several reasons. Systematic samples tend to be easier to draw and execute. The
researcher does not have to jump backward and forward through the sampling frame to draw
the members to be sampled. A systematic sample may spread the members selected for
measurement more evenly across the entire population than simple random sampling.
m
Therefore, in some cases, systematic sampling may be more representative of the population
and more precise.
o
One of the most attractive aspects of systematic sampling is that this method can allow
the researcher to draw a probability sample without complete prior knowledge of the sampling
.c
frame. For example, a survey of visitors to the County’s publications desk could be conducted
by sampling every 10th visitor after randomly selecting the first through 10th visitor as the
ts
starting point. By conducting the sample in this manner, it would not be necessary for the
researcher to obtain a comprehensive list of visitors prior to drawing the sample.
n
As with other types of cluster sampling, systematic sampling is as precise as simple
ra
random sampling if the members contained in the clusters are as heterogeneous as the
population. If this assumption is not valid, then systematic sampling will be less precise than
simple random sampling. In conducting systematic sampling, it is also essential that the
pi
researcher does not introduce bias into the sample by selecting an inappropriate sampling
interval. For instance, when conducting a sample of financial records, or other items that
follow a calendar schedule, the researcher would not want to select “7” as the sampling interval
as
because the sample would then be comprised of observations that were all on the same day of
the week. Day-of-the-week influences may cause contamination of the sample, giving the
g
researcher biased results.

.c
(v) Multi-Stage Sampling. Multi-stage sampling is like cluster sampling, but involves se-
lecting a sample within each chosen cluster, rather than including all units in the cluster.
Thus, multi-stage sampling involves selecting a sample in at least two stages. In the first
w
stage, large groups or clusters are selected. These clusters are designed to contain more popu-
lation units than are required for the final sample.
w
In the second stage, population units are chosen from selected clusters to derive a final
sample. If more than two stages are used, the process of choosing population units within
w
clusters continues until the final sample is achieved.

An example of multi-stage sampling is where, firstly, electoral sub-divisions (clusters)
are sampled from a city or state. Secondly, blocks of houses are selected from within the
electoral sub-divisions and, thirdly, individual houses are selected from within the selected
blocks of houses.
The advantages of multi-stage sampling are convenience, economy and efficiency. Multi-
stage sampling does not require a complete list of members in the target population, which
greatly reduces sample preparation cost. The list of members is required only for those clusters
used in the final stage. The main disadvantage of multi-stage sampling is the same as for
cluster sampling : lower accuracy due to higher sampling error.
www.cgaspirants.com
www.cgaspirants.com
3.73 PARAMETERS OF STATISTICS
The statistical constants of the population such as mean, the variance etc. are known as the
parameters. The statistical concepts of the sample from the members of the sample to estimate
the parameters of the population from which the sample has been drawn is known as statistic.
Population mean and variance are denoted by μ and σ2, while those of the samples are
given by x , s2.
3.74 STANDARD ERROR
The standard deviation of the sampling distribution of a statistic is known as the standard
error (S.E.). It plays an important role in the theory of large samples and it forms a basis of
m
t − E(t)
the testing of hypothesis. If t is any statistic, for large sample. z = is normally distrib-
S. E. (t)
o
uted with mean 0 and variance unity.
.c
3.75 TEST OF SIGNIFICANCE
ts
An important aspect of the sampling theory is to study the test of significance which will
enable us to decide, on the basis of the results of the sample, whether
n
(i) the deviation between the observed sample statistic and the hypothetical parameter
ra
value or
(ii) the deviation between two sample statistics
pi
is significant or might be attributed due to chance or the fluctuations of the sampling.
3.76 TESTING OF STATISTICAL HYPOTHESIS

as
Step 1. Null hypothesis: [U.P.T.U. (MCA) 2007]

g
For applying the tests of significance, we first set up a hypothesis which is a definite
statement about the population parameter called Null Hypothesis. It is denoted by H0.
.c
Null hypothesis is the hypothesis which is tested for possible rejection under the as-
w
sumption that it is true. First, we set up H0 in clear terms.

Step 2. Alternative hypothesis:
w
Any hypothesis which is complementary to the null hypothesis (H0) is called an alterna-
tive hypothesis. It is denoted by H1.
w
For example, if we want to test the null hypothesis that the population has a specified
mean μ0 then we have
H0 : μ = μ0
then the alternative hypothesis will be
(i) H1 : μ ≠ μ0 (Two tailed alternative hypothesis)
(ii) H1 : μ > μ0 (right tailed alternative hypothesis (or) single tailed)
(iii) H1 : μ < μ0 (left tailed alternative hypothesis (or) single tailed)
Hence alternative hypothesis helps to know whether the test is two tailed or one tailed.
Therefore, we set up H1 for this decision.
www.cgaspirants.com
www.cgaspirants.com
Step 3. Level of significance: [U.P.T.U. (MCA) 2008, 2007]

The probability of the value of the variate falling in the critical region is known as level
of significance. A region corresponding to a statistic t in the sample space S which amounts to
rejection of the null hypothesis H0 is called as critical region or region of rejection while
which amounts to acceptance of H0 is called acceptance region. The probability α that a
random value of the statistic t belongs to the critical region is known as the level of
significance.
P (t ∈ w/H0) = α
i.e., the level of significance is the size of the type I error (refer art. 3.77) or the maximum
producer’s risk.
We select the appropriate level of significance in advance depending on the reliability of
m
the estimates.
Step 4. Test statistic (or test criterion): We compute the test statistic z under the
o
t − E(t)
null hypothesis. For larger samples corresponding to the statistic t, the variable z = is
S.E.(t)
.c
normally distributed with mean 0 and variance 1. The value of z given above under the null
hypothesis is known as test statistic.
ts
Step 5. Conclusion: We compare the computed value of z with the critical value zα at
level of significance (α). The critical value of zα of the test statistic at level of significance α for
a two tailed test is given by
n
ra
p(|z| > zα) = α …(1)

i.e., zα is the value of z so that the total area of the critical region on both tails is α. Since
the normal curve is symmetrical, from equation (1), we get
pi
p(z > zα) + p(z < – zα) = α ; i.e., 2p(z > zα) = α ; i.e., p(z > zα) = α/2
i.e., the area of each tail is α/2.
g as
.c
w
w
w
The critical value zα is that value such that the area to the right of zα is α/2 and the area
to the left of – zα is α/2.
In the case of one tailed test,
p(z > zα) = α if it is right tailed ; p(z < – zα) = α if it is left tailed.
The critical value of z for a single tailed test (right or left) at level of significance α is
same as the critical value of z for two tailed test at level of significance 2α.
Using the equation, also using the normal tables, the critical value of z at different
levels of significance (α) for both single tailed and two tailed test are calculated and listed
below. The equations are
p(| z | > zα ) = α ; p(z > zα ) = α ; p(z < – zα ) = α
www.cgaspirants.com
www.cgaspirants.com
Level of significance
1% (0.01) 5% (0.05) 10% (0.1)
Two tailed test | zα | = 2.58 | zα | = 1.96 | zα | = 1.645
Right tailed test zα = 2.33 zα = 1.645 zα = 1.28
Left tailed test zα = – 2.33 zα = – 1.645 zα = – 1.28
If |z| > zα, we reject H0 and conclude that there is significant difference. If |z| < zα, we
accept H0 and conclude that there is no significant difference.
3.77 ERRORS IN SAMPLING
m
The main aim of the sampling theory is to draw a valid conclusion about the population
parameters on the basis of the sample results. In doing this we may commit the following two
o
types of errors:
.c
Type I. Error. [U.P.T.U. (MCA) 2008, 2007]
When H0 is true, we may reject it.
P(Reject H0 when it is true) = P(Reject H0/H0) = α
ts
α is called the size of the type I error also referred to as producer’s risk.
n
Type II. Error. When H0 is wrong, we may accept it P(Accept H0 when it is wrong) =
ra
P(Accept H0/H1) = β . β is called the size of the type II error, also referred to as consumer’s
risk.
pi
Note. The values of the test statistic which separates the critical region and acceptance region are
called the critical values or significant values. This value is dependent on (i) the level of significance
used and (ii) the alternative hypothesis, whether it is one tailed or two tailed.
as
3.78 TEST OF SIGNIFICANCE OF SMALL SAMPLES

g
When the size of the sample is less than 30, then the sample is called small sample. For such
.c
sample it will not be possible for us to assume that the random sampling distribution of a
statistic is approximately normal and the values given by the sample data are sufficiently
w
close to the population values and can be used in their place for the calculation of the standard
error of the estimate.
w
3.79 STUDENT’S t-DISTRIBUTION (t-Test)

w
[G.B.T.U. (MBA) 2011 ; G.B.T.U. (MCA) 2010]
This t-distribution is used when sample size is ≤ 30 and the population standard deviation is
unknown.
t-statistic is defined as
x−μ Σ( x − x ) 2
t= where, S=
S/ n n−1
x is the mean of sample, μ is population mean. S is the standard deviation of population
and n is sample size.
www.cgaspirants.com
www.cgaspirants.com
If the standard deviation of the sample ‘s’ is given then t-statistic is defined as
x−μ
t=
s/ n − 1
Note. The relation between s and S is ns2 = (n – 1)S2.
3.79.1. The t-Table

The t-table given at the end is the probability integral of t-distribution. The t-distribution has
different values for each degrees of freedom and when the degrees of freedom are infinitely
large, the t-distribution is equivalent to normal distribution and the probabilities shown in
the normal distribution tables are applicable.
m
3.79.2. Applications of t-Distribution [G.B.T.U. (MBA) 2011]
Some of the applications of t-distribution are given below:
o
1. To test if the sample mean ( x ) differs significantly from the hypothetical value μ of
.c
the population mean.
2. To test the significance between two sample means.
3.79.3. Critical Value of t

n ts
3. To test the significance of observed partial and multiple correlation coefficients.
ra
The critical value or significant value of t at level of significance α, degrees of freedom γ for two
tailed test is given by
P[| t | > tγ (α)] = α
pi
P[| t | ≤ tγ (α)] = 1 – α
The significant value of t at level of significance α, for a single tailed test can be got from
as
those of two tailed test by referring to the values at 2α.
3.80 TEST I : t-TEST OF SIGNIFICANCE OF THE MEAN OF A RANDOM SAMPLE

g
.c
To test whether the mean of a sample drawn from a normal population deviates significantly
from a stated value when variance of the population is unknown.
w
H0 : There is no significant difference between the sample mean x and the population
w
mean μ i.e., we use the statistic
x −μ Σ( x − x ) 2
w
t= where S=
S/ n n−1
with degree of freedom n – 1.
At given level of significance α and degrees of freedom (n – 1), we refer to t-table tα (two
tailed or one tailed). If calculated t value is such that | t | < tα, the null hypothesis is accepted.
If | t | > tα, H0 is rejected.
3.80.1. Fiducial Limits of Population Mean

If tα is the value of t at level of significance α at (n – 1) degrees of freedom then,
x−μ
< tα for acceptance of H0.
S/ n
www.cgaspirants.com
www.cgaspirants.com
x – tα S/ n < μ < x + tα S/ n
95% confidence limits (level of significance 5%) are x ± t0.05 S/ n .
99% confidence limits (level of significance 1%) are x ± t0.01 S/ n .
EXAMPLES
Example 1. A random sample of size 16 has 53 as mean. The sum of squares of the
deviation from mean is 135. Can this sample be regarded as taken from the population having
56 as mean? Obtain 95% and 99% confidence limits of the mean of the population.
Sol. Null hypothesis, H0 : There is no significant difference between the sample mean
m
and hypothetical population mean i.e., μ = 56.
Alternative hypothesis, H1 : μ ≠ 56 (Two tailed test)
o
x−μ
Test statistic. Under H0, test statistic is t =
S/ n
.c
x
Given: = 53, μ = 56, n = 16, Σ ( x − x ) 2 = 135
S=
Σ( x − x ) 2
x−μ
n−1
n =
53 − 56
135
15
=3 ts
ra
t= = =–4
S/ n 3/ 16
|t|=4
pi
d.fv. = 16 – 1 = 15.
Conclusion. Since | t | = 4 > t0.05 = 2.13 i.e., the calculated value of t is more than the
as
tabulated value, the null hypothesis is rejected. Hence, the sample mean has not come from a
population having 56 as mean.
95% confidence limits of the population mean
g
S 3
.c
= x± t0.05 = 53 ± (2.13) = 51.4025, 54.5975

n 16
99% confidence limits of the population mean
w
S 3
= x± t0.01, = 53 ± (2.95) = 50.7875, 55.2125.
w
n 16
Example 2. The lifetime of electric bulbs for a random sample of 10 from a large con-
w
signment gave the following data:

Item 1 2 3 4 5 6 7 8 9 10
Life in ’000 hrs. 4.2 4.6 3.9 4.1 5.2 3.8 3.9 4.3 4.4 5.6
Can we accept the hypothesis that the average lifetime of bulb is 4000 hrs?
Sol. Null hypothesis: H0: There is no significant difference in the sample mean and
population mean. i.e., μ = 4000 hrs.
Alternative hypothesis: μ ≠ 4000 hrs (Two tailed test)
x−μ
Test statistic: Under H0, the test statistic is t =
S/ n
www.cgaspirants.com
www.cgaspirants.com
x 4.2 4.6 3.9 4.1 5.2 3.8 3.9 4.3 4.4 5.6
x– x – 0.2 0.2 – 0.5 – 0.3 0.8 – 0.6 – 0.5 – 0.1 0 1.2
(x − x) 2 0.04 0.04 0.25 0.09 0.64 0.36 0.25 0.01 0 1.44
Σx 44 2
x = = = 4.4, Σ ( x − x ) = 3.12
n 10
Σ( x − x ) 2
S= = 0.589
n−1
x−μ 4.4 − 4
t= = = 2.123
S/ n 0.589FG IJ
m
10H K
For γ = 9, t0.05 = 2.26.
o
Conclusion. Since the calculated value of t is less than the tabulated value of t at 5%
.c
level of significance. ∴ The null hypothesis μ = 4000 hrs is accepted i.e., the average lifetime
of bulbs could be 4000 hrs.
that it is a random sample from a normal population with mean 45 units.

n ts
Example 3. A sample of 20 items has mean 42 units and S.D. 5 units. Test the hypothesis
Sol. Null hypothesis: H0 : There is no significant difference between the sample mean
and the population mean. i.e., μ = 45 units
ra
Alternative hypothesis, H1: μ ≠ 45 (Two tailed test)

Given: n = 20, x = 42, s = 5 ; γ = 19 d.f.
pi
Test statistic: Under H0, the test statistic is

x−μ 42 − 45
t= = = – 2.615
as
s/ n − 1 5/ 19
∴ | t | = 2.615
g
The tabulated value of t at 5% level for 19 d.f. is t0.05 = 2.09.

.c
Conclusion. Since the calculated value | t | is greater than the tabulated value of t at
5% level of significance, the null hypothesis H0 is rejected. i.e., there is significant difference
w
between the sample mean and population mean.

i.e., the sample could not have come from this population.
w
Example 4. The 9 items of a sample have the following values

45, 47, 50, 52, 48, 47, 49, 53, 51.
w
Does the mean of these values differ significantly from the assumed mean 47.5?
Σx
Sol. Here, n = 9, μ = 47.5, x = = 49.1
n
x 45 47 50 52 48 47 49 53 51
x– x – 4.1 – 2.1 0.9 2.9 – 1.1 – 2.1 – .1 3.9 1.9
(x – x )2 16.81 4.41 .81 8.41 1.21 4.41 .01 15.21 3.61
Σ(x – x )2 = 54.89,
Σ (x − x)2
S2 = = 6.86
n−1
∴ S = 2.619
www.cgaspirants.com
www.cgaspirants.com
Null hypothesis:
H0 : μ = 47.5
i.e., there is no significant difference between the sample and population means.
Alternative hypothesis:
H1 : μ ≠ 47.5
Hence we apply two-tailed test.
x − μ 49.1 − 47.5
t= = = 1.8327
S/ n (2.619 / 9 )
t0.05 = 2.31 for γ = 8
Conclusion: Since |t|calculated < ttabulated at 5% level of significance, the null hypothesis
m
H0 is accepted i.e., there is no significant difference between their means.
o
ASSIGNMENT
.c
1. Ten individuals are chosen at random from a normal population of students and their marks are
that mean mark of the population of students is 66.

n ts
found to be 63, 63, 66, 67, 68, 69, 70, 70, 71, 71. In the light of these data, discuss the suggestion
2. The following values gives the lengths of 12 samples of Egyptian cotton taken from a consign-
ment : 48, 46, 49, 46, 52, 45, 43, 47, 47, 46, 45, 50. Test if the mean length of the consignment can
ra
be taken as 46.
3. A sample of 18 items has a mean 24 units and standard deviation 3 units. Test the hypothesis
pi
that it is a random sample from a normal population with mean 27 units.

4. A random sample of 10 boys had the I.Q.’s 70, 120, 110, 101, 88, 83, 95, 98, 107 and 100. Do these
as
data support the assumption of a population mean I.Q. of 160?
3.81 TEST II : t-TEST FOR DIFFERENCE OF MEANS OF TWO SMALL SAMPLES

g
(from a Normal Population)

.c
This test is used to test whether the two samples x1, x2, ......, xn1 , y1, y2, ......, yn2 of sizes n1, n2
w
have been drawn from two normal populations with mean μ1 and μ2 respectively under the
assumption that the population variance are equal. (σ1 = σ2 = σ).
w
H0 : The samples have been drawn from the normal population with means μ1 and μ2
i.e., H0 : μ1 ≠ μ2.
w
Let x , y be their means of the two samples.

( x − y)
Under this H0 the test statistic t is given by t =
1 1
S +
n1 n2
Degree of freedom is n1 + n2 – 2.
n1s12 + n2 s22
Note 1. If the two sample’s standard deviations s1, s2 are given then we have S2 = .
n1 + n2 − 2
Σ ( x1 − x1)2 + Σ ( x2 − x2 )2
Note 2. If s1, s2 are not given then S2 = .
n1 + n2 − 2
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Two samples of sodium vapour bulbs were tested for length of life and the
following results were got:
Size Sample mean Sample S.D.
Type I 8 1234 hrs 36 hrs
Type II 7 1036 hrs 40 hrs
Is the difference in the means significant to generalise that Type I is superior to Type II
regarding length of life?
m
Sol. Null hypothesis,
H0 : μ1 = μ2 i.e., two types of bulbs have same lifetime.
o
Alternative hypothesis,
H1 : μ1 > μ2 i.e., type I is superior to Type II.
.c
Hence we use right tailed test.
∴
S2 =
S = 40.7317
n1 + n2 − 2
n =
8+7−2 ts
n1s12 + n2 s2 2 8(36) 2 + 7(40) 2
= 1659.076
ra
Test statistic: Under H0, the test statistic t is given by
x1 − x2 1234 − 1036
t= = = 18.1480
pi
1 1 1 1
S + 40.7317 +
n1 n2 8 7
as
t0.05 at d.f. γ = n1 + n2 – 2 = 13 is 1.77.

Conclusion. Since calculated | t | > ttabulated at 5% level of significance, H0 is rejected.
g
∴ Type I is definitely superior to Type II.

.c
Example 2. Samples of sizes 10 and 14 were taken from two normal populations with
S.D. 3.5 and 5.2. The sample means were found to be 20.3 and 18.6. Test whether the means of
w
the two populations are the same at 5% level.

Sol. We have, x1 = 20.3, x2 = 18.6, n1 = 10, n2 = 14, s1 = 3.5, s2 = 5.2
w
n1s12 + n2 s22
S2 = = 22.775
w
n1 + n2 − 2
∴ S = 4.772
Null hypothesis:
H0 : μ1 = μ2 i.e., the means of the two populations are the same.
Alternative hypothesis:
H1 : μ1 ≠ μ2
x1 − x2 20.3 − 18.6
t= = = 0.8604
1 1 1 1
S + 4.772 +
n1 n2 10 14
The tabulated value of t at 5% level of significance for 22 d.f. is t0.05 = 2.0739
www.cgaspirants.com
www.cgaspirants.com
Conclusion:
Since t = 0.8604 < t0.05, the null hypothesis H0 is accepted; i.e., there is no significant
difference between their means.
Example 3. The height of 6 randomly chosen sailors in inches are 63, 65, 68, 69, 71 and
72. Those of 9 randomly chosen soldiers are 61, 62, 65, 66, 69, 70, 71, 72 and 73. Test whether
the sailors are on the average taller than soldiers.
Sol. Let X1 and X2 be the two samples denoting the heights of sailors and soldiers.
n1 = 6, n2 = 9
Null hypothesis, H0 : μ1 = μ2.
i.e., the mean of both the population are the same.
m
Alternative hypothesis, H1 : μ1 > μ2 (one tailed test)
Calculation of two sample means:
o
X1 63 65 68 69 71 72
.c
X1 – X1 –5 –3 0 1 3 4
(X1 − X1 ) 2 25
n9 0 ts 1 9 16
ΣX 1
ra
X1 = = 68 ; Σ (X 1 − X 1 ) 2 = 60
n1
pi
X2 61 62 65 66 69 70 71 72 73
as
X2 – X 2 – 6.66 – 5.66 – 2.66 1.66 1.34 2.34 3.34 4.34 5.34
(X 2 − X 2 ) 2 44.36 32.035 7.0756 2.7556 1.7956 5.4756 11.1556 18.8356 28.5156

g
.c
ΣX 2
X2 = = 67.66 ; Σ (X 2 − X 2 ) 2 = 152.0002
n2
w
1
S2 = [ Σ (X 1 − X 1 ) 2 + Σ (X 2 − X 2 ) 2 ] = 16.3077
w
n1 + n2 − 2
∴ S = 4.038
w
Test statistic:
X1 − X2 68 − 67.666
Under H0, t= = = 0.1569
1 1 1 1
S + 4.038 +
n1 n2 6 9
The value of t at 5% level of significance for 13 d.f. is 1.77. (d.f. = n1 + n2 – 2)
Conclusion. Since tcalculated < t0.05 = 1.77, the null hypothesis H0 is accepted.
i.e., there is no significant difference between their average.
i.e., the sailors are not on the average taller than the soldiers.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. The mean life of 10 electric motors was found to be 1450 hrs with S.D. of 423 hrs. A second
sample of 17 motors chosen from a different batch showed a mean life of 1280 hrs with a S.D. of
398 hrs. Is there a significant difference between means of the two samples?
2. The marks obtained by a group of 9 regular course students and another group of 11 part time
course students in a test are given below:
Regular: 56 62 63 54 60 51 67 69 58
Part time: 62 70 71 62 60 56 75 64 72 68 66
Examine whether the marks obtained by regular students and part time students differ signifi-
cantly at 5% and 1% level of significance.
m
3. A group of 5 patients treated with the medicine A weigh 42, 39, 48, 60 and 41 kgs. A second group
of 7 patients from the same hospital treated with medicine B weigh 38, 42, 56, 64, 68, 69 and
o
62 kgs. Do you agree with the claim that medicine B increases the weight significantly? It is
given that the value of t at 10% level of significance for 10 degree of freedom is 1.81.
.c
[G.B.T.U. (B. Pharm.) 2010]
4. Two independent samples of sizes 7 and 9 have the following values:
Sample A: 10
Sample B: 10
12
13
10
15
13
12
14
10
n 11
14
10
11 12
Test whether the difference between the mean is significant.
11 ts
ra
5. The average no. of articles produced by two machines per day are 200 and 250 with standard
deviations 20 and 25 respectively on the basis of records of 25 days production. Can you regard
both the machines equally efficient at 5% level of significance?
pi
3.82 CHI-SQUARE (χ2) TEST [G.B.T.U. 2010 ; G.B.T.U. MCA (SUM) 2010]
as
When a coin is tossed 200 times, the theoretical considerations lead us to expect 100 heads
and 100 tails. But in practice, these results are rarely achieved. The quantity χ2 (a Greek
letter, pronounced as chi-square) describes the magnitude of discrepancy between theory and
g
observation. If χ2 = 0, the observed and expected frequencies completely coincide. The greater
.c
the discrepancy between the observed and expected frequencies, the greater is the value of χ2.
Thus χ2 affords a measure of the correspondence between theory and observation.
w
If Oi (i = 1, 2, ......, n) is a set of observed (experimental) frequencies and Ei (i = 1, 2, ......, n)

is the corresponding set of expected (theoretical or hypothetical) frequencies, then, χ2 is de-
w
fined as
w
n
L (O − Ei ) 2 OP
χ2 = ∑ MMN
i=1
i
Ei PQ
where ΣOi = ΣEi = N (total frequency) and degrees of freedom (d.f.) = (n – 1).
Note. (i) If χ2 = 0, the observed and theoretical frequencies agree exactly.
(ii) If χ2 > 0 they do not agree exactly.
3.83 DEGREES OF FREEDOM
While comparing the calculated value of χ2 with the tabular value, we have to determine the
degrees of freedom.
www.cgaspirants.com
www.cgaspirants.com
If we have to choose any four numbers whose sum is 50, we can exercise our independ-
ent choice for any three numbers only, the fourth being 50 minus the total of the three num-
bers selected. Thus, though we were to choose any four numbers, our choice was reduced to
three because of one condition imposed. There was only one restraint on our freedom and our
degrees of freedom were 4 – 1 = 3. If two restrictions are imposed, our freedom to choose will
be further curtailed and degrees of freedom will be 4 – 2 = 2.
In general, the number of degrees of freedom is the total number of observations less
the number of independent constraints imposed on the observations. Degrees of freedom (d.f.)
are usually denoted by ν.
Thus, ν = n – k, where k is the number of independent constraints in a set of data of n
observations.
Note. (i) For a p × q contingency table (p columns and q rows), ν = (p – 1) (q – 1)
m
(ii) In the case of a contingency table, the expected frequency of any class
Total of rows in which it occurs × Total of columns in which it occurs
o
=
Total number of observations
.c
3.84 APPLICATIONS OF CHI-SQUARE TEST
n ts
χ2 test is one of the simplest and the most general test known. It is applicable to a very large
number of problems in practice which can be summed up under the following heads:
(i) as a test of goodness of fit.
ra
(ii) as a test of independence of attributes.

(iii) as a test of homogeneity of independent estimates of the population variance.
pi
(iv) as a test of the hypothetical value of the population variance σ2.

(v) as a list to the homogeneity of independent estimates of the population correlation
as
coefficient.
3.85 CONDITIONS FOR APPLYING χ2 TEST

g
χ2-test is an approximate test for large values of n. For the validity of chi-square test of goodness
.c
of fit between theory and experiment, the following conditions must be satisfied.
w
(a) The sample observations should be independent.

(b) The constraints on the cell frequencies, if any, should be linear e.g., Σni = Σλi
w
or ΣOi = ΣEi.
(c) N, the total number of frequencies should be reasonably large. It is difficult to say
w
what constitutes largeness, but as an arbitrary figure, we may say that N should be atleast
50, however, few the cells.
(d) No theoretical cell-frequency should be small. Here again, it is difficult to say what
constitutes smallness, but 5 should be regarded as the very minimum and 10 is better. If
small theoretical frequencies occur (i.e., < 10), the difficulty is overcome by grouping two or
more classes together before calculating (O – E). It is important to remember that the
number of degrees of freedom is determined with the number of classes after
regrouping.
Note 1. If any one of the theoretical frequency is less than 5, then we apply a corrected given by F Yates,
which is usually known as ‘Yates correction for continuity’, we add 0.5 to the cell frequency which is less
than 5 and adjust the remaining cell frequency suitably so that the marginal total is not changed.
www.cgaspirants.com
www.cgaspirants.com
Note 2. It may be noted that the χ2-test depends only on the set of observed and expected frequencies
and on degrees of freedom (d.f.). It does not make any assumption regarding the parent population from
which the observations are taken. Since χ2 does not involve any population parameters, it is termed as
a statistic and the test is known as Non-parametric test or Distribution-free test.
3.86 THE χ2 DISTRIBUTION
For large sample sizes, the sampling distribution of χ2 can be closely approximated by a
continuous curve known as the chi-square distribution. The probability function of χ 2
distribution is given by
2
f(χ2) = c(χ2)(ν/2–1) e − x / 2
where e = 2.71828, ν = number of degrees of freedom ; c = a constant depending only on ν.
m
Symbolically, the degrees of freedom are denoted by the symbol ν or by d.f. and are
obtained by the rule ν = n – k, where k refers to the number of independent constraints.
o
In general, when we fit a binomial distribution the number of degrees of freedom is one
less than the number of classes ; when we fit a Poisson distribution the degrees of freedom are
.c
2 less than the number of classes, because we use the total frequency and the arithmetic mean
to get the parameter of the Poisson distribution. When we fit a normal curve the number of
total frequency, mean and standard deviation.

n ts
degrees of freedom are 3 less than the number of classes, because in this fitting we use the
If the data is given in a series of “n” numbers then degrees of freedom = n – 1.

ra
In the case of Binomial distribution d.f. = n – 1
In the case of Poisson distribution d.f. = n – 2
In the case of Normal distribution d.f. = n – 3.
pi
3.87 χ2 TEST AS A TEST OF GOODNESS OF FIT

as
χ2 test enables us to ascertain how well the theoretical distributions such as Binomial, Poisson
or Normal etc. fit empirical distributions, i.e., distributions obtained from sample data. If the
g
calculated value of χ2 is less than the tabular value at a specified level (generally 5%) of
.c
significance, the fit is considered to be good i.e., the divergence between actual and expected
frequencies is attributed to fluctuations of simple sampling. If the calculated value of χ2 is
w
greater than the tabular value, the fit is considered to be poor.

w
EXAMPLES
w
Example 1. In experiments on pea breeding, the following frequencies of seeds were

obtained:
Round and Wrinkled and Round and Wrinkled and Total

yellow yellow green green
315 101 108 32 556
Theory predicts that the frequencies should be in proportions 9 : 3 : 3 : 1. Examine the

correspondence between theory and experiment.
Sol. Null hypothesis:
H0 : The experimental result support the theory i.e., there is no significant difference
between the observed and theoretical frequency.
www.cgaspirants.com
www.cgaspirants.com
Under H0, The theoretical (expected) frequencies can be calculated as follows:

556 × 9 556 × 3
E1 = = 312.75 E2 = = 104.25
16 16
556 × 3 556 × 1
E3 = = 104.25 E4 = = 34.75
16 16
To calculate the value of χ2:
Observed frequency Oi 315 101 108 32

Expected Frequency Ei 312.75 104.25 104.25 34.75
m
(Oi − Ei ) 2
0.016187 0.101319 0.134892 0.217626
Ei
o
L (O − Ei )2 OP = 0.470024
∑ MMN i
.c
χ2 =
Ei PQ
Tabular value of χ2 at 5% level of significance for n – 1 = 3 d.f. is 7.815 i.e., χ20.05 = 7.815.
hence H0 is accepted. Therefore, the experimental results support the theory.

n ts
Conclusion: Since the calculated value of χ2 is less than that of the tabulated value,
Example 2. The following table gives the number of accidents that took place in an
ra
industry during various days of the week. Test if accidents are uniformly distributed over the
week.
pi
Day Mon Tue Wed Thu Fri Sat

as
No. of accidents 14 18 12 11 15 14
Sol. Null hypothesis H0: The accidents are uniformly distributed over the week.
g
84
.c
Under this H0, the expected frequencies of the accidents on each of these days = = 14
6
w
Observed frequency Oi 14 18 12 11 15 14
w
Expected frequency Ei 14 14 14 14 14 14
(Oi – Ei)2 0 16 4 9 1 0
w
(O L − Ei ) 2 OP
Σ (O i − E i ) 2 30
χ2 = ∑ MMN i
Ei
=
PQ Ei
=
14
= 2.1428.
Tabular value of χ2 at 5% level for (6 – 1 = 5 d.f.) is 11.09.

Conclusion: Since the calculated value of χ2 is less than the tabulated value, H0 is
accepted i.e., the accidents are uniformly distributed over the week.
Example 3. A die is thrown 276 times and the results of these throws are given below:
No. appeared on the die 1 2 3 4 5 6

Frequency 40 32 29 59 57 59
Test whether the die is biased or not.
www.cgaspirants.com
www.cgaspirants.com
Sol. Null hypothesis H0 : Die is unbiased.

276
Under this H0, the expected frequencies for each digit is = 46.
6
To find the value of χ2
Oi 40 32 29 59 57 59
Ei 46 46 46 46 46 46
(Oi – Ei)2 36 196 289 169 121 169
L (O − Ei ) 2 OP Σ(O − E ) 2
980
χ2 = ∑ MMN i
Ei PQ = E
i
i
i
=
46
= 21.30.
m
Tabulated value of χ2
at 5% level of significance for (6 – 1 = 5) d.f. is 11.09.
Conclusion. Since the calculated value of χ2 = 21.30 > 11.07 the tabulated value, H0 is
o
rejected. i.e., die is not unbiased or die is biased.
Example 4. Records taken of the number of male and female births in 800 families
.c
having four children are as follows: [U.P.T.U. (MCA) 2009]
No. of male births

No. of female births
No. of families 32
0
4
n 1
3
178
ts
290
2
2
3
1
236
4
0
64
ra
Test whether the data are consistent with the hypothesis that the Binomial law holds
and the chance of male birth is equal to that of female birth, namely p = q = 1/2.
pi
Sol. Null hypothesis H0: The data are consistent with the hypothesis of equal
probability for male and female births. i.e., p = q = 1/2.
as
We use Binomial distribution to calculate theoretical frequency given by:

N(r) = N × P(X = r) = N × nCr prqn–r
g
where N is the total frequency, N(r) is the number of families with r male children,
p and q are probabilities of male and female births respectively, n is the number of
.c
children.
FG 1IJ
w
4
N(0) = 800 × 4C0
H 2K = 50, N(1) = 200, N(2) = 300, N(3) = 200 and N(4) = 50
w
Observed frequency Oi 32 178 290 236 64

w
Expected frequency Ei 50 200 300 200 50
(Oi – Ei)2 324 484 100 1296 196
(Oi − Ei ) 2
6.48 2.42 0.333 6.48 3.92
Ei
L(O − Ei ) 2 OP = 19.633.
χ2 = ∑ MMN i
Ei PQ
Tabulated value of χ2 at 5% level of significance for 5 – 1 = 4 d.f. is 9.49.
Conclusion. Since the calculated value of χ2 is greater than the tabulated value, H0 is
rejected. i.e., the data are not consistent with the hypothesis that the Binomial law holds and
that the chance of a male birth is not equal to that of a female birth.
www.cgaspirants.com
www.cgaspirants.com
Example 5. The theory predicts the proportion of beans in the four groups, G1, G2, G3,
G4 should be in the ratio 9 : 3 : 3 : 1. In an experiment with 1600 beans the numbers in the four
groups were 882, 313, 287 and 118. Does the experimental result support the theory?
Sol. Null hypothesis H0: The experimental result support the theory. i.e., there is no
significant difference between the observed and theoretical frequency.
Under H0, the theoretical frequency can be calculated as follows:
1600 × 9 1600 × 3
E(G1) = = 900; E(G2) = = 300;
16 16
1600 × 3 1600 × 1
E(G3) = = 300; E(G4) = = 100
16 16
m
To calculate the value of χ2.
o
Observed frequency Oi 882 313 287 118
.c
Expected frequency Ei 900 300 300 100
(Oi − Ei ) 2
Ei
n 0.36
ts
0.5633 0.5633 3.24
L (O − Ei ) 2 OP = 4.7266.
ra
χ2 = ∑ MMN i
Ei PQ
pi
Tablular value of χ2 at 5% level of significance for 3 d.f. is 7.815.

Conclusion: Since the calculated value of χ2 is less than that of the tabulated value,
as
hence H0 is accepted. i.e., the experimental results support the theory.

Example 6. The following table shows the distribution of digits in numbers chosen at
random from a telephone directory:
g
.c
Digits 0 1 2 3 4 5 6 7 8 9
Frequency 1026 1107 997 966 1075 933 1107 972 964 853
w
Test whether the digits may be taken to occur equally frequently in the directory.
w
[G.B.T.U. (MCA) 2011]

Sol. Null hypothesis:
w
H0 : The digits taken in the directory occur equally frequently i.e., there is no signifi-
cant difference between the observed and expected frequency.
10000
Under H0, the expected frequency = = 1000
10
Calculation of χ2
Oi 1026 1107 997 966 1075 933 1107 972 964 853
Ei 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000
(Oi – Ei)2 676 11449 9 1156 5625 4489 11449 784 1296 21609
www.cgaspirants.com
www.cgaspirants.com
L (O OP
− Ei ) 2 58542
χ2 = ∑ MMN i
Ei
=
PQ 1000
= 58.542
The tabulated value of χ2 at 5% level of significance for 9 d.f. is 16.919.

Conclusion: Since χ2calculated > χ2tabulated, H0 is rejected i.e., there is significant difference
between the observed and theoretical frequencies. Therefore, the digits taken in the directory
donot occur equally frequently.
Example 7. When the first proof of 392 pages of a book of 1200 pages were read, the
distribution of printing mistakes were found to be as follows:
m
No. of mistakes in a page (x): 0 1 2 3 4 5 6
No. of pages (f): 275 72 30 7 5 2 1
o
Fit a poisson distribution to the above data and test the goodness of fit.
.c
Sol. Null Hypothesis, H0 : Poisson distribution is a good fit to the data.
Mean (λ) =
∑ fx = 189 = 0.4821
∑ f 392
n ts
ra
The frequency of x mistakes per page is given by the poisson law as follows:
N(x) = N . P(x)
pi
LM e −0.4821
(0.4821) x OP = 242.05(0.4821) x
as
= 392 ;0≤x≤6
N x! Q x!
Under H0, expected frequencies are,

g
.c
N(0) = 242.05, N(1) = 116.69, N(2) = 28.13, N(3) = 4.52

N(4) = 0.54, N(5) = 0.052, N(6) = 0.0042
w
The χ2-table is as follows:

w
Mistakes Observed frequency Expected frequency (Oi – Ei)2

(Oi − Ei )2
per page (Oi) (Ei)
w
Ei
(x) (correct to one
place of decimal)
0 275 242.1 1082.41 4.471

1 72 116.7 1998.09 17.121
2 30 28.1 3.61 0.128
3 7 U| 4.5 U|
4 5 0.5
5 2
V| 15 0.1
V| 5.1 98.01 19.217
6 1 W 0 W
Total 392 392 40.937
www.cgaspirants.com
www.cgaspirants.com
L (O − Ei ) 2 OP = 40.937
χ 2cal. = ∑ MMN i
Ei PQ
d.f. = 7 – 1 – 1 – 3 = 2
One d.f. is lost because of linear constraint ΣOi = ΣEi. One d.f. is lost because the
parameter λ has been estimated from the given data and is then used for computing the
expected frequencies. 3 d.f. are lost because of grouping the last four expected cell frequencies
which were less than 5.
Tabulated value of χ2 for 2 d.f. at 5% level of significance is 5.991.
2 2
Conclusion : Since χ cal. > χ tab. , the null hypothesis is rejected at 5% level of significance.
m
Hence, we conclude that poisson distribution is not a good fit to the given data.
Example 8. Fit a Poisson distribution to the following data and best the goodness of fit :
o
x: 0 1 2 3 4
f: 109 65 22 3 1
.c
Sol. Null hypothesis, H0: Poisson distribution is a good fit to the data.
Mean (λ) =
Σ fx 122
Σf
n =
200
= 0.61
ts
e − .61 (.61) x (108.67) (0.61) x
N(x) = N . P(x) = (200) =
ra
x! x!
Under H0, expected frequencies are
pi
N(0) = 108.67 ≈ 109, N(1) = 66.29 ≈ 66, N(2) = 20.22 ≈ 20

N(3) = 4.11 ≈ 4, N(4) = 0.63 ≈ 1
as
The χ2-table is as follows:

g
(Oi − Ei ) 2
x Oi Ei (Oi – Ei)2
Ei
.c
0 109 109 0 0
w
1 65 66 1 0.01515
2 22 20 4 0.2
w
3 3 UV 4 4 UV 5 1 0.2
4 1 W 1 W
w
Total 200 200 0.41515
L (O − Ei ) 2 OP
2
χ cal . = ∑ MMN i
Ei PQ = 0.41515
d.f. = 5 – 1 – 1 – 1 = 2
Tabulated value of χ2 for 2 d.f. at 5% level of significance is 5.991.
2 2
Conclusion: Since χ cal. < χ tab. , the null hypothesis H0 is accepted at 5% level of
significance. Hence we conclude that Poisson distribution is a good fit to the given data.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. A sample analysis of examination results of 500 students, it was found that 220 students have
failed, 170 have secured a third class, 90 have secured a second class and the rest, a first class.
Do these figures support the general belief that above categories are in the ratio 4 : 3 : 2 : 1
respectively ? (The tabular value of χ2 for d.f. 3 at 5% level of significance is 7.81).
[U.P.T.U. (MBA) 2009]
2. What is χ2–test? [G.B.T.U. 2010 ; G.B.T.U. MCA (C.O.) 2010]
A die is thrown 90 times with the following results:
Face: 1 2 3 4 5 6 Total
Frequency: 10 12 16 14 18 20 90
m
Use χ2-test to test whether these data are consistent with the hypothesis that die is unbiased.
Given χ20.05 = 11.07 for 5 degrees of freedom. [U.P.T.U. (MCA) 2007]
3. A survey of 320 families with 5 children shows the following distribution:
o
No. of boys 5 boys 4 boys 3 boys 2 boys 1 boy 0 boy Total
.c
& girls: & 0 girl & 1 girl & 2 girls & 3 girls & 4 girls & 5 girls
No. of
families: 18 56 110
ts
88 40 8 320
Given that values of χ2 for 5 degrees of freedom are 11.1 and 15.1 at 0.05 and 0.01 significance
n
level respectively, test the hypothesis that male and female births are equally probable.
(G.B.T.U. 2010)
ra
4. A chemical extraction plant processes sea water to collect sodium chloride and magnesium. It is
known that sea water contains sodium chloride, magnesium and other elements in the ratio
pi
62 : 4 : 34. A sample of 200 tonnes of sea water has resulted in 130 tonnes of sodium chloride and
6 tonnes of magnesium. Are these data consistent with the known composition of sea water at 5%
level of significance? (Given that the tabular value of χ2 is 5.991 for 2 degree of freedom).
as
[U.P.T.U. MCA (C.O.) 2008]

5. The demand for a particular spare part in a factory was found to vary from day-to-day. In a
g
sample study, the following information was obtained:

Days: Mon Tue Wed Thurs Fri Sat
.c
No. of parts demanded: 1124 1125 1110 1120 1126 1115

Test the hypothesis that the number of parts demanded does not depend on the day of the week.
w
[Given. The values of chi-square significance at 5, 6, 7 d.f. are respectively 11.07, 12.59, 14.07 at
5% level of significance] (G.B.T.U. 2011)
w
6. The sales in a supermarket during a week are given below. Test the hypothesis that the sales
donot depend on the day of the week using a significant level of 0.05.
w
Days Mon Tue Wed Thurs Fri Sat
Sales 65 54 60 56 71 84
(in 1000 `)
7. 4 coins were tossed at a time and this operation is repeated 160 times. It is found that 4 heads
occur 6 times, 3 heads occur 43 times, 2 heads occur 69 times, one head occur 34 times. Discuss
whether the coin may be regarded as unbiased?
8. 200 digits are chosen at random from a set of tables. The frequencies of the digits were :
Digits: 0 1 2 3 4 5 6 7 8 9
Frequency: 18 19 23 21 16 25 22 20 21 15
www.cgaspirants.com
www.cgaspirants.com
Use χ2-test to assess the correctness of the hypothesis that the digits were distributed in equal
numbers in the table, given that the value of χ2 are respectively 16.9, 18.3 and 19.7 for 9, 10 and
11 degrees of freedom at 5% level of significance.
9. A genetical law says that children having one parent of blood group M and the other parent of
blood group N will always be one of the three blood groups M, MN, N and that the average no. of
children in these groups will be in the ratio 1 : 2 : 1. The report on an experiment states as
follows:
“Of 162 children having one M parent and one N parent, 28.4% were found to be of group M, 42%
of group MN and the rest of the group N.” Do the data in the report conform to the expected
genetic ratio 1 : 2 : 1?
10. Every clinical thermometer is classified into one of the four categories A, B, C and D on the basis
of inspection and test. From past experience, it is knwon that thermometers produced by a certain
m
manufacturer are distributed among the four categories in the following proportions:
Category: A B C D
Proportion: 0.87 0.09 0.03 0.01
o
A new lot of 1336 thermometers is submitted by the manufacturer for inspection and test and the
.c
following distribution into four categories results :
Category: A B C D
No. of the rmometers reported: 1188 91
ts
47
Does this new lot of thermometers differ from the previous experience with regards to proportion
of thermometers in each category?
n 10
ra
11. Test for goodness of fit of a poisson distribution at 5% level of significance to the following frequency
distribution:
(i) x: 0 1 2 3 4 5 6 7 8
pi
f: 52 151 130 102 45 12 5 1 2

[Hint. Group the last three frequencies]
as
(ii) x: 0 1 2 3 4 5 6 7 8 9 10 11 12 13
f: 3 15 47 76 68 74 46 39 15 9 5 2 0 1
g
[Hint. Group the first two and last four frequencies]

.c
(iii) x: 0 1 2 3 4 5
f: 275 138 75 7 4 1
w
[Hint. Club the last three frequencies]

(iv) x: 0 1 2 3 4
w
f: 419 352 154 56 19

12. (i) Fit a binomial distribution to the data and test for goodness of fit at 5% level of significance.
w
x: 0 1 2 3 4 5
y: 38 144 342 287 164 25
(ii) A random number table of 250 digits showed the following distribution of digits 0, 1, 2, ..., 9.
Digit: 0 1 2 3 4 5 6 7 8 9
Observed: 17 31 29 18 14 20 35 30 20 36
Frequency
Expected: 25 25 25 25 25 25 25 25 25 25
Frequency
Does the observed distribution differ significantly from expected distributions using a significance
level of 0.01? Given that χ20.99 for 9 degrees of freedom is 21.7. [G.B.T.U. MCA (SUM) 2010]
www.cgaspirants.com
www.cgaspirants.com
Answers
1. No 2. Yes
3. H0 accepted at 1% level of significance and rejected at 5% level of significance
4. χ2 = 1.025, Yes 5. H0 accepted at 5% level 6. H0 accepted
7. Coin is unbiased 8. H0 accepted at 5% level 9. H0 accepted at 5% level
10. H0 rejected at 5% level
11. (i) H0 accepted at 5% level, (ii) H0 accepted at 5% level, (iii) H0 rejected at 5% level, not a good fit.
(iv) H0 accepted at 5% level
12. (i) H0 accepted at 5% level, provides a good fit (ii) Yes.
3.88 χ2 TEST AS A TEST OF INDEPENDENCE
m
With the help of χ2 test, we can find whether or not, two attributes are associated. We take the
o
null hypothesis that there is no association between the attributes under study, i.e., we assume
that the two attributes are independent. If the calculated value of χ2 is less than the
.c
table value at a specified level (generally 5%) of significance, the hypothesis holds good, i.e.,
the attributes are independent and do not bear any association. On the other hand, if the
ts
calculated value of χ2 is greater than the table value at a specified level of significance, we say
that the results of the experiment do not support the hypothesis. In other words, the attributes
n
are associated. Thus a very useful application of χ2 test is to investigate the relationship
between trials or attributes which can be classified into two or more categories.
ra
The sample data set out into two-way table, called contingency table.
Let us consider two attributes A and B divided into r classes A1, A2, A3, ......, Ar and B
pi
divided into s classes B1, B2, B3, ......, Bs. If (Ai), (Bj) represents the number of persons possess-
ing the attributes Ai, Bj respectively, (i = 1, 2, ......, r, j = 1, 2, ......, s) and (Ai Bj) represent the
as
r s
number of persons possessing attributes Ai and Bj. Also we have ∑ Ai = ∑B j = N where N
g
i=1 j=1
is the total frequency. The contingency table for r × s is given as follows:

.c
A A1 A2 A3 ...Ar Total
w
B
w
B1 (A1B1) (A2B1) (A3B1) ......(ArB1) B1

B2 (A1B2) (A2B2) (A3B2) ......(ArB2) B2
w
B3 (A1B3) (A2B3) (A3B3) ......(ArB3) B3

...... ...... ...... ...... ...... ......
...... ...... ...... ...... ...... ......
Bs (A1Bs) (A2Bs) (A3Bs) ......(ArBs) (Bs)
Total (A1) (A2) (A3) ......(Ar) N
H0 : Both the attributes are independent. i.e., A and B are independent under the null
hypothesis, we calculate the expected frequency as follows:
(A i )
P(Ai) = Probability that a person possesses the attribute Ai = i = 1, 2, ......, r
N
www.cgaspirants.com
www.cgaspirants.com
(B j )
P(Bj) = Probability that a person possesses the attribute Bj =
N
(A i B j )
P(AiBj) = Probability that a person possesses both attributes Ai and Bj =
N
If (AiBj)0 is the expected number of persons possessing both the attributes Ai and Bj
(AiBj)0 = NP(AiBj) = NP(Ai)(Bj)
( A i ) ( B j ) ( A i )( B j )
=N = (∵ A and B are independent)
N N N
L [(A B ) − (A B ) ] OP
m
r s 2
Hence χ2 = ∑ ∑ MM i j i
PQ
j 0
N
i=1 j=1
(A B ) i j 0
o
which is distributed as a χ2 variate with (r – 1)(s – 1) degrees of freedom.
.c
a/b 2
ts
Note 1. For a 2 × 2 contingency table where the frequencies are , χ can be calculated from independent
c/d
(a + b + c + d)(ad − bc)2
n
frequencies as χ2 = .
(a + b)(c + d)(b + d)(a + c)
ra
Note 2. If the contingency table is not 2 × 2, then the above formula for calculating χ2 can’t be used.
(A i )(B j )
pi
Hence, we have another formula for calculating the expected frequency (AiBj)0 =
N
Product of column total and row total

as
i.e., expected frequency in each cell is = .

whole total
a|b ad − bc
g
Note 3. If is the 2 × 2 contingency table with two attributes, Q = is called the coefficient
c|d ad + bc
.c
a c
of association. If the attributes are independent then= .
b d
w
Note 4. Yates’s Correction. In a 2 × 2 table, if the frequencies of a cell is small, we make Yates’s
w
1
correction to make χ2 continuous. Decrease by those cell frequencies which are greater than expected
2
w
1
frequencies, and increase by those which are less than expectation. This will not affect the marginal
2
columns. This correction is known as Yates’s correction to continuity. After Yates’s correction
FG 1 IJ 2
χ2 =
H
N bc − ad −
2
N
K when ad – bc < 0
(a + c)(b + d)(c + d)(a + b)
FG 1 IJ 2
and 2
χ =
H
N ad − bc −
2
N
K when ad – bc > 0.
(a + c)(b + d)(c + d)(a + b)
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. What are the expected frequencies of 2 × 2 contingency tables given below:
a b 2 10
(i) (ii)
c d 6 6
Sol. Observed frequencies Expected frequencies
( a + c )( a + b) (b + d)(a + b)
(i) a b a+b
a+b+c+d
m
a+b+c+d
c d c+d →
o
(a + c)(c + d) (b + d)(c + d)
a+c b+d a+b+c+d=N
.c
a+b+c+d a+b+c+d
Observed frequencies Expected frequencies
(ii) 2 10
n 12 ts 8 × 12
24
=4
16 × 12
24
=8
6 6 12 →
ra
8 × 12 16 × 12
8 16 24 =4 =8
24 24
pi
Example 2. From the following table regarding the colour of eyes of father and son, test
if the colour of son’s eye is associated with that of the father.
as
Eye colour of son

g
Light Not light

.c
Eye colour of father Light 471 51

w
Not light 148 230
Sol. Null hypothesis H0: The colour of son’s eye is not associated with that of the
w
father. i.e., they are independent.

w
Under H0, we calculate

Product of column total and row total
the expected frequency in each cell =
whole total
www.cgaspirants.com
www.cgaspirants.com
Expected frequencies are:
Eye colour
of son Light Not light Total
Eye colour
of father
619 × 522 289 × 522
Light = 359.02 = 167.62 522
900 900
619 × 378 289 × 378

Not light = 259.98 = 121.38 378
900 900
m
Total 619 289 900
(471 − 359.02) 2 (51 − 167.62) 2 (148 − 259.98) 2 (230 − 121.38) 2
o
χ2 = + + + = 261.498.
359.02 167.62 259.98 121.38
.c
Tabulated value of χ2 at 5% level for 1 d.f. is 3.841.
Conclusion. Since the calculated value of χ2 > tabulated value of χ2, H0 is rejected.
ts
They are dependent i.e., the colour of son’s eye is associated with that of the father.
Example 3. The following table gives the number of good and bad parts produced by
each of the three shifts in a factory:
n
ra
Good parts Bad parts Total

Day shift 960 40 1000
pi
Evening shift 940 50 990

as
Night shift 950 45 995

Total 2850 135 2985
g
Test whether or not the production of bad parts is independent of the shift on which they
.c
were produced.
Sol. Null hypothesis H0. The production of bad parts is independent of the shift on
w
which they were produced. i.e., the two attributes, production and shifts are independent.
2 L [(A B ) − (A B )]
3 2 OP
∑ ∑ MM
w
i j 0 i j
Under H0, χ2 =
N (A B )
i=1 j=1 i j 0 PQ
w
Calculation of expected frequencies

Let A and B be the two attributes namely production and shifts. A is divided into two
classes A1, A2 and B is divided into three classes B1, B2, B3.
(A 1 )(B 1 ) (2850) × (1000)
(A1B1)0 = = = 954.77
N 2985
(A 1 )(B 2 ) (2850) × (990)
(A1B2)0 = = = 945.226
N 2985
( A 1 )(B 3 ) (2850) × (995)
(A1B3)0 = = = 950
N 2985
(A 2 )(B 1 ) (135) × (1000)
(A2B1)0 = = = 45.27
N 2985
www.cgaspirants.com
www.cgaspirants.com
(A 2 )(B 2 ) (135) × (990)

(A2B2)0 = = = 44.773
N 2985
( A 2 )(B 3 ) (135) × (995)
(A2B3)0 = = = 45.
N 2985
To calculate the value of χ2
Class Oi Ei (Oi – Ei)2 (Oi – Ei)2/Ei
(A1B1) 960 954.77 27.3529 0.02864
(A1B2) 940 945.226 27.3110 0.02889
m
(A1B3) 950 950 0 0
o
(A2B1) 40 45.27 27.7729 0.61349
.c
(A2B2) 50 44.773 27.3215 0.61022
(A2B3) 45 45 0 0
n ts
The tabulated value of χ2 at 5% level of significance for 2 degrees of freedom (r – 1)
1.28126
ra
(s – 1) is 5.991.
Conclusion: Since the calculated value of χ2 is less than the tabulated value, we accept
H0. i.e., the production of bad parts is independent of the shift on which they were produced.
pi
Example 4. From the following data, find whether hair colour and sex are associated.
as
Colour Fair Red Medium Dark Black Total

Sex
g
Boys 592 849 504 119 36 2100

.c
Girls 544 677 451 97 14 1783

w
Total 1136 1526 955 216 50 3883

w
Sol. Null hypothesis H0. The two attributes hair colour and sex are not associated.
w
i.e., they are independent.

Let A and B be the attributes hair colour and sex respectively. A is divides into 5 classes
(r = 5). B is divided into 2 classes (s = 2).
∴ Degrees of freedom = (r – 1)(s – 1) = (5 – 1)(2 – 1) = 4
5 2
[( A i B j ) 0 − ( A i B j )]2
Under H0, we calculate χ2 = ∑∑
i=1 j=1
(A i B j ) 0
To calculate the expected frequency (Ai Bj)0 as follows:
(A 1 )(B 1 ) 1136 × 2100

(A1B1)0 = = = 614.37
N 3883
www.cgaspirants.com
www.cgaspirants.com
(A 1 )(B 2 ) 1136 × 1783

(A1B2)0 = = = 521.629
N 3883
(A 2 )(B 1 ) 1526 × 2100
(A2B1)0 = = = 852.289
N 3883
(A 2 )(B 2 ) 1526 × 1783
(A2B2)0 = = = 700.71
N 3883
(A 3 )(B 1 ) 955 × 2100

(A3B1)0 = = = 516.482
N 3883
m
(A 3 )(B 2 ) 955 × 1783
(A3B2)0 = = = 483.517
N 3883
o
(A 4 )(B 1 ) 216 × 2100
(A4B1)0 = = = 116.816
.c
N 3883
(A 4 )(B 2 ) 216 × 1783
(A4B2)0 =
(A5B1)0 =
N
=
=
n3883
(A 5 )(B 1 ) 50 × 2100
ts
= 99.183
= 27.04
ra
N 3883
(A 5 )(B 2 ) 50 × 1783
(A5B2)0 = = = 22.959
pi
N 3883
Calculation of χ 2
as
(Oi − Ei ) 2
Class Oi Ei (Oi – Ei )2
Ei
g
.c
A1B1 592 614.37 500.416 0.8145
A1B2 544 521.629 500.462 0.959

w
A2B1 849 852.289 10.8175 0.0127

w
A2B2 677 700.71 562.1641 0.8023

w
A3B1 504 516.482 155.800 0.3016
A3B2 451 438.517 155.825 0.3553
A4B1 119 116.816 4.7698 0.0408
A4B2 97 99.183 4.7654 0.0480
A5B1 36 27.04 80.2816 2.9689
A5B2 14 22.959 80.2636 3.495
9.79975
www.cgaspirants.com
www.cgaspirants.com
χ 2cal. = 9.799.
Tabular value of χ2 at 5% level of significance for 4 d.f. is 9.488.
Conclusion: Since the calculated value of χ2 < tabulated value, H0 is rejected. i.e., the
two attributes are not independent. i.e., the hair colour and sex are associated.
Example 5. Can vaccination be regarded as preventive measure of small pox as evi-
denced by the following data of 1482 persons exposed to small pox in a locality. 368 in all were
attacked of these 1482 persons and 343 were vaccinated and of these only 35 were attacked.
Sol. For the given data we form the contingency table. Let the two attributes be vacci-
nation and exposed to small pox. Each attributes is divided into two classes.
m
Vaccination A
Disease small Vaccinated Not Total
pox B
o
Attacked 35 333 368
.c
Not 308 806 1114
Total
n 343
ts 1139
Null hypothesis H0. The two attributes are independent i.e., vaccination can’t be
1482
regarded as preventive measure of small pox.

ra
Degrees of freedom ν = (r – 1)(s – 1) = (2 – 1)(2 – 1) = 1

pi
2 2
[( A i B j ) 0 − ( A i B j )]2
Under H0, χ2 = ∑∑
i=1 j=1
(A i B j ) 0
as
Calculation of expected frequency

(A 1 )(B 1 ) 343 × 368
g
(A1B1)0 = = = 85.1713
N 1482
.c
(A 1 )(B 2 ) 343 × 1114

(A1B2)0 = = = 257.828
N 1482
w
(A 2 )(B 1 ) 1139 × 368

(A2B1)0 = = = 282.828
N 1482
w
(A 2 )(B 2 ) 1139 × 1114

(A2B2)0 = = = 856.171
w
N 1482
Calculation of χ 2
Class Oi Ei (Oi – Ei)2 (Oi − Ei ) 2

Ei
(A1B1) 35 85.1713 2517.159 29.554
(A1B2) 308 257.828 2517.229 8.1728
(A2B1) 333 282.828 2517.2295 7.5592
(A2B2) 806 856.171 2517.1292 2.9399
48.2261
www.cgaspirants.com
www.cgaspirants.com
Calculated value of χ2 = 48.2261.

Tabulated value of χ2 at 5% level of significance for 1 d.f. is 3.841.
Conclusion. Since the calculated value of χ2 > tabulated value, H0 is rejected.
i.e., the two attributes are not independent. i.e., the vaccination can be regarded as
preventive measure of small pox.
Example 6. To test the effectiveness of inoculation against cholera, the following table
was obtained:
Attacked Not attacked Total
Inoculated 30 160 190
m
Not inoculated 140 460 600
Total 170 620 790
o
(The figures represent the number of persons.)
.c
Use χ2-test to defend or refute the statement that the inoculation prevents attack from
cholera. (U.P.T.U. 2009)
Under H0, we calculate the expected frequencies as:

n ts
Sol. Null hypothesis H0: The inoculation does not prevent attack from cholera.
Attacked Not attacked

ra
190 × 170 190 × 620

Inoculated = 40.886 = 149.11
pi
790 790
600 × 170 600 × 620

as
Not inoculated = 129.11 = 470.89

790 790
Calculation of χ2
g
.c
Oi 30 160 140 460

Ei 40.886 149.11 129.11 470.89
w
(Oi − Ei ) 2
w
2.898 0.795 0.918 0.252

Ei
w
(O i − E i ) 2
χ 2cal. = ∑ Ei
= 4.863
Tabulated value of χ2 at 5% level of significance for 1 d.f. is 3.841.

Conclusion: Since χ 2cal. > χ 2tab. at 5% level of significance, null hypothesis H0 is rejected.
Hence we defend the statement that inoculation prevents attack from cholera.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. In a locality 100 persons were randomly selected and asked about their educational achieve-
ments. The results are given below:
Education
Middle High school College
Sex Male 10 15 25
Female 25 10 15
Based on this information can you say that the education depends on sex.
m
2. The following data is collected on two characters:
o
Smokers Non smokers
Literate 83 57
.c
Illiterate 45 68
3.
n ts
Based on this information can you say that there is no relation between habit of smoking and
literacy.
500 students at school were graded according to their intelligences and economic conditions of
ra
their homes. Examine whether there is any association between economic condition and intelli-
gence, from the following data:
pi
Economic conditions Intelligence

Good Bad
as
Rich 85 75
Poor 165 175
g
4. In an experiment on the immunisation of goats from anthrox, the following results were
.c
obtained. Derive your inferences on the efficiency of the vaccine.

w
Died anthrox Survived

Inoculated with vaccine 2 10
w
Not inoculated 6 6
w
5. By using χ2-test, find out whether there is any association between income level and type of
schooling:
Income Public School Govt. School
Low 200 400
High 1000 400
2
(Given for degree of freedom 1, χ0.05 = 3.84) [U.P.T.U. 2008, G.B.T.U. (MBA) 2011]
6. Examine by any suitable method, whether the nature of area is related to voting preference in
the election for which the data are tabulated below:
www.cgaspirants.com
www.cgaspirants.com
Votes for Area A B Total
Rural 620 480 1100
Urban 380 520 900
Total 1000 1000 2000
(U.P.T.U. 2006)
7. The groups of 100 people each were taken for testing the use of a vaccine. 15 persons contracted
the disease out of the inoculated persons, while 25 contracted the disease in the other group. Test
the efficiency of the vaccine using Chi-square test. (The value of χ2 for one degree of freedom at
5% level of significance is 3.84).
m
Answers
1. Yes 2. No 3. No 4. Not effective
o
5. Yes 6. Yes 7. Not associated.
.c
TEST YOUR KNOWLEDGE
1.
n ts
The fourth moment about the mean of a frequency distribution is 24. What must be the value of
standard deviation in order that the distribution be platykurtic? (M.T.U. 2012)
ra
2. Two events A and B have probabilities 0.25 and 0.50 respectively. The probability that both
events A and B occurs in 0.14. Find the probability that neither A nor B occurs. (M.T.U. 2013)
pi
Hint. P(A ∩ B) = P(A ∪ B) = 1 − P( A ∪ B)
3. (i) Define the coefficients of skewness and kurtosis. (U.P.T.U. 2014)

as
(ii) Define skewness, coefficient of skewness, kurtosis and coefficient of kurtosis. (M.T.U. 2013)
(iii) Write a short note on skewness. [(M.T.U. (B.Pharma) 2011]
g
(iv) What is meant by skewness? How is it measured? [(M.T.U. (MBA) 2012]

.c
4. What is the total probability theorem? (U.P.T.U. 2014)

5. The first four central moments of a distribution are 0, 2.5, 0.7 and 18.75. Comment on the kurtosis
w
of the distribution. (M.T.U. 2013)

6. Find the moment generating function of Poisson distribution. (M.T.U. 2013)
w
9
7. Find the parameters p and q of the binomial distribution whose mean is 9 and variance is .
4
w
(M.T.U. 2012)
8. If the sum of the mean and variance of a binomial distribution of 5 trials is 9 , find P(X ≥ 1).
5
(M.T.U. 2013)
9. It has been found that 2% of the tools produced by a certain machine are defective. What is the
probability that in a shipment of 400 such tools, 3 or more will be defective? (M.T.U. 2013)
10. If P(X = 0) = P(X = 1) = k in a Poisson distribution, then what is k?
11. For a Poisson variate X if P(X = 1) = P(X = 2), then find P(X = 4).
12. Find the total area under the curve of p.d.f. of a normal curve.
13. If for a Poisson distribution, P(2) = P(3), then what is its probability function?
www.cgaspirants.com
www.cgaspirants.com
14. Find the parameters of a binomial distribution with mean = 8 and variance = 4.
15. If X is a normal variate with mean 30 and S.D. 5, find the probabilities that
(i) 26 ≤ X ≤ 40 (ii) X ≥ 45 and (iii) |X – 30 | > 5
Answers
t 1 3
2. 0.39 5. β2 = 3, Mesokurtic 6. Mx (t) = eλ( e − 1)
7. q = ,p=
4 4
1 2
8. 0.67232 9. 0.9862 10. 11.
e 3 e2
e−3 (3) x 1
12. 1 13. 14. n = 16, p =
m
x! 2
15. (i) 0.7653 (ii) 0.00135 (iii) 0.3174.
o
.c
n ts
ra
pi
g as
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
UNIT 4
Numerical Techniques–I
m
4.1 SOLUTION OF ALGEBRAIC AND TRANSCENDENTAL EQUATIONS
o
Consider the equation of the form f(x) = 0
.c
If f(x) is a quadratic, cubic or biquadratic expression then algebraic formulae are available
for expressing the roots. But when f(x) is a polynomial of higher degree or an expression involving
are not available.

n ts
transcendental functions e.g., 1 + cos x – 5x, x tan x – cosh x, e–x – sin x etc., algebraic methods
Here, we shall describe some numerical methods for the solution of f(x) = 0, where f(x) is
ra
algebraic or transcendental.
4.2 ORDER OF CONVERGENCE OF ITERATIVE METHODS

pi
Convergence of an iterative method is judged by the order at which the error between succes-
as
sive approximations to the root decreases.

An iterative method is said to be kth order convergent if k is the largest positive real
number such that
g
ei + 1
.c
lim ≤A
i→∞ ei k
w
where A is a non-zero finite number called asymptotic error constant and it depends on
derivative of f(x) at an approximate root x. ei and ei+1 are the errors in successive approxima-
w
tions. kth order convergence gives us idea in each iteration, the no. of significant digits in each
approximation increases k times.
w
The error in any step is proportional to the kth power of the error in the previous step.
4.3 CONVERGENCE OF A SEQUENCE
A sequence < xn > of successive approximations of a root x = α of the equation f(x) = 0 is said to
converge to x = α with order p ≥ 1 iff
| x n + 1 – α | ≤ c | x n – α | p, n ≥ 0
c being some constant greater than zero.
Particularly, if | xn + 1 – α | = c | xn – α |, n ≥ 0, 0 < c < 1 then convergence is called
geometric. Also, if p = 1 and 0 < c < 1 then convergence is called linear or of first order.
Constant c is called the rate of linear convergence. Convergence is rapid or slow according
as c is near 0 or 1.
379
www.cgaspirants.com
www.cgaspirants.com
Using induction, condition for linear convergence can be simplified to the form
| xn – α | ≤ cn | x0 – α |, n ≥ 0, 0 < c < 1.
4.4 BISECTION METHOD
This method is based on the repeated application Y
of intermediate value property.
Let the function f(x) be continuous between

a and b. For definiteness, let f(a) be (–)ve and f (b)
)
f(x
be (+)ve. Then the first approximation to the root is
=
y
1
x1 = (a + b). f(b)
2
m
If f(x1) = 0, then x1 is a root of f(x) = 0 other- x2
a
wise, the root lies between a and x1 or x1 and b
o
O x1 b X
according as f(x1) is (+) ve or (–) ve. Then, we bisect f(a)
.c
the interval as before and continue the process until
the root is found to desired accuracy.
n ts
In the above figure, f(x1) is (+)ve so that the root lies between a and x1. Then second
1
approximation to the root is x2 = (a + x1). If f (x2) is (–)ve ; the root lies between x1 and x2.
2
1
ra
Then the third approximation to the root is x3 = (x1 + x2) and so on.
2
We observe that this method uses only the end points of the interval [an, bn] for which
pi
f(an) · f(bn) < 0 and not the values of f(x) at these end points to obtain the next approximation
to the root.
as
The number of iterations required may be determined from the relation

|b − a|
≤ε
g
2n
.c
log e |b − a|− log e ε

or n≥
log e 2
w
where ε is the permissible error.

w
This method requires a large number of iterations to achieve a reasonable degree of

accuracy for the root. It requires one function evaluation for each iteration. This method is
w
also called as Bolzano method or Interval halving method.
4.5 PROVE THAT BISECTION METHOD ALWAYS CONVERGES
Let [ pn, qn] be the interval at nth step of bisection, having a root of the equation f(x) = 0. Let xn
be the nth approximation for the root. Then initially p1 = a and q1 = b.
FG p 1 + q1 IJ
⇒ x1 = first approximation = H 2 K
⇒ p 1 < x 1 < q1
Now either the root lies in [a, x1] or in [x1, b].
∴ either [p2, q2] = [p1, x1] or [p2, q2] = [x1, q1]
www.cgaspirants.com
www.cgaspirants.com
NUMERICAL TECHNIQUES–I 381
⇒ either p 2 = p 1 , q2 = x 1 or p 2 = x 1 , q2 = q1
⇒ p 1 ≤ p 2, q 2 ≤ q 1
p2 + q2
Also, x2 = so that p2 < x2 < q2
2
Continuing this way, we obtain that at nth step,
p + qn
xn = n , pn < x n < qn
2
and p1 ≤ p2 ≤ ...... ≤ pn and q1 ≥ q2 ≥ ...... ≥ qn
∴ < p1, p2, ......, pn, ...... > is bounded non-decreasing sequence bounded by b and
< q1, q2, ......, qn, ...... > is a bounded non-increasing sequence of numbers bounded by a.
Hence, both these sequences converge.
m
Let, lim pn = p and lim qn = q.
n→∞ n→∞
o
Now since length of the interval is decreasing at every step, we get that
lim (qn – pn) = 0 q=p
.c
⇒
n→∞
Also, p n < xn < q n

⇒
⇒
lim pn ≤ lim xn ≤ lim qn
n
p ≤ lim xn ≤ q
ts
⇒ lim xn = p = q …(1)
ra
Further since a root lies in [pn, qn], we shall have

f( pn) · f(qn) < 0
pi
⇒ 0 ≥ lim [f(pn)] . f(qn)]

n→∞
as
⇒ 0 ≥ f(p) . f(q)
⇒ 0 ≥ [f(p)]2
g
But [ f( p)]2 ≥ 0 being a square

.c
∴ we get f( p) = 0
∴ p is a root of f( x) = 0 …(2)
w
From (1) and (2), we see that <xn> converges necessarily to a root of equation f(x) = 0
The method is not rapidly converging but it is useful in the sense that it converges
w
surely.
w
EXAMPLES
Example 1. Find the real root of the equation x log10 x = 1.2 by bisection method correct
to four decimal places.
Sol. f(x) = x log10 x – 1.2 ...(1)
Since, f(2.74) = – .000563 i.e., (–)ve
and f(2.75) = .0081649 i.e., (+)ve
Hence, a root lies between 2.74 and 2.75.
∴ First approximation to the root is
2.74 + 2.75
x1 = = 2.745
2
www.cgaspirants.com
www.cgaspirants.com
Now, f(x1) = f(2.745) = .003798 i.e., (+)ve

Hence, root lies between 2.74 and 2.745.
∴ Second approximation to the root is
2.74 + 2.745
x2 = = 2.7425
2
Now, f(x2) = f(2.7425) = .001617 i.e., (+)ve
∴ Third approximation to the root is
2.74 + 2.7425
x3 = = 2.74125
2
Now, f(x3) = f(2.74125) = .0005267 i.e., (+)ve
m
∴ Fourth approximation to the root is
o
2.74 + 2.74125
x4 = = 2.740625
.c
2
Now, f(x4) = f(2.740625) = – .00001839 i.e., (–)ve.
∴ Fifth approximation to the root is
n
2.740625 + 2.74125
ts
x5 = = 2.7409375
ra
2
Now, f(x5) = f(2.7409375) = .000254 i.e., (+)ve
pi
∴ Sixth approximation to the root is

2.740625 + 2.7409375
as
x6 = = 2.74078125
2
Now, f(x6) = f(2.74078125) = .0001178 i.e., (+)ve
g

∴ Seventh approximation to the root is
.c
2.740625 + 2.74078125
x7 = = 2.740703125
w
2
Since, x6 and x7 are same up to four decimal places hence the approximate real root is
w
2.7407.
Example 2. Find a positive real root of x – cos x = 0 by bisection method, correct up to
w
4 decimal places between 0 and 1.

Sol. Let f(x) = x – cos x
f(0.73) = (–)ve and f(0.74) = (+)ve
Hence, the root lies between 0.73 and 0.74.
First approximation to the root is
0.73 + 0.74
x1 = = 0.735
2
Now, f(0.735) = (–)ve
Second approximation to the root is
0.73 + 0.74
x2 = = 0.7375
2
www.cgaspirants.com
www.cgaspirants.com
Now, f(0.7375) = (–)ve

Third approximation to the root is
0.7375 + 0.74
x3 = = 0.73875
2
Now, f(0.73875) = (–)ve
Fourth approximation to the root is
1
x4 = (0.73875 + 0.74) = 0.739375
2
m
Now, f(x4) = f(0.739375) = (+)ve
o
Fifth approximation to the root is
.c
1
x5 = (0.73875 + 0.739375) = 0.7390625
2
Now, f(0.7390625) = (–)ve
ts
Hence, the root lies between 0.7390625 and 0.739375
n
Sixth approximation to the root is
ra
1
x6 = (0.7390625 + 0.739375) = 0.73921875
2
pi
Now, f(0.73921875) = (+)ve

Seventh approximation to the root is
as
1
x7 = (0.7390625 + 0.73921875) = 0.73914
2
g
Now, f(0.73914) = (+)ve

.c

w
Eighth approximate to the root is

1
w
x8 = (0.7390625 + 0.73914) = 0.73910

2
Since x7 and x8 are same up to four decimal places hence the approximate real root is
w
0.7391.
Example 3. Perform five iterations of bisection method to obtain the smallest positive
root of equation f(x) ≡ x3 – 5x + 1 = 0. [G.B.T.U 2011; G.B.T.U. (C.O.) 2011]
Sol. 3
f(x) = x – 5x + 1 ...(1)
Since, f(0.2016) = 0.0001935 i.e., (+)ve
and f(0.2017) = – 0.0002943 i.e., (–)ve
First approximation to the root is
0.2016 + .2017
x1 = = 0.20165
2
www.cgaspirants.com
www.cgaspirants.com
Now, f(x1) = – 0.00005036 i.e., (–)ve

0.2016 + 0.20165
x2 = = 0.201625
2
Now, f(x2) = 0.00007159 i.e., (+)ve
Hence, root lies between 0.201625 and .20165.
0.201625 + 0.20165
x3 = = 0.2016375
2
m
Now, f(x3) = 0.00001061 i.e., (+)ve
o
Fourth approximation to the root is
.c
0.2016375 + 0.20165
x4 = = 0.20164375
Now,
2
ts
f(x4) = – 0.00001987 i.e., (–)ve
n
ra
0.2016375 + 0.20164375
x5 = = 0.201640625
2
pi
Hence, after performing five iterations, the smallest positive root of the given equation
is 0.20164 correct to five decimal places.
as
Example 4. Find a real root of x3 – x = 1 between 1 and 2 by bisection method. Compute

five iterations.
g
Sol. Here, f(x) = x3 – x – 1 ...(1)

.c
Since, f(1.324) = – 0.00306 i.e., (–)ve

and f(1.325) = 0.00120 i.e., (+)ve
w

w
1.324 + 1.325
x1 = = 1.3245
w
2
Now, f(x1) = – 0.000929 i.e., (–)ve
Hence, root lies between 1.3245 and 1.325
1.3245 + 1.325
x2 = = 1.32475
2
Now, f(x2) = 0.000136 i.e., (+)ve
1.3245 + 1.32475
x3 = = 1.324625
2
www.cgaspirants.com
www.cgaspirants.com
Now, f(x3) = – 0.000396 i.e., (–)ve

1.324625 + 1.32475
x4 = = 1.3246875
2
Now, f(x4) = – .0001298 i.e., (–)ve
1.3246875 + 1.32475
x5 = = 1.32471875
2
m
Hence, the real root of the given equation is 1.324 correct to three decimal places after
computing five iterations.
o
Example 5. Use bisection method to find out the positive square root of 30 correct to
four decimal places.
.c
Sol. Let f(x) = x2 – 30 ...(1)
and
Since, f(5.477) = – .00247 i.e., (–)ve
f(5.478) = .00848 i.e., (+)ve
n
ts
ra
5.477 + 5.478
x1 = = 5.4775
2
pi
Now, f(x1) = .003 i.e., (+)ve

as

5.477 + 5.4775
g
x2 = = 5.47725
2
.c
Now, f(x2) = .00026 i.e., (+)ve

w
∴ Third approximation to the root is

w
5.477 + 5.47725
x3 = = 5.477125
2
w
Now, f(x3) = – .0011 i.e., (–)ve

5.477125 + 5.47725
x4 = = 5.4771875
2
Since, x3 and x4 are same up to four decimal places hence the positive square root of 30
correct to 4 decimal places is 5.4771.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. (i) Transcendental equation is given as

f(x) = 2x – x – 3
Calculate f(x) for x = – 4, – 3, – 2, – 1, 0, 1, 2, 3, 4 and compute, between which integers values
roots are lying.
(ii) The equation x2 – 2x – 3cos x = 0 is given. Locate the smallest root in magnitude in an interval
of length one unit.
2. Find a real root of ex = 3x by Bisection method.
3. The smallest positive root of the equation f(x) = x4 – 3x2 + x – 10 = 0 is to be obtained.
(i) Find an interval of unit length which contains this root.
m
(ii) Perform two iterations of the Bisection method.
4. Find real root lying in interval (1, 2) up to four decimal places for the equation x6 – x4 – x3 – 1 = 0 by
o
Bisection method.
5. Find a real root of cos x – xex = 0 correct to three decimal places by Bisection method.
.c
(U.P.T.U. 2008)
6. The negative root of the smallest magnitude of the equation f(x) = 3x3 + 10x2 + 10x + 7 = 0 is to be
obtained.
ts
(i) Find an interval of unit length which contains this root.
n
(ii) Perform two iterations of Bisection method.
ra
7. Compute the root of f(x) = sin 10x + cos 3x using Bisection method. The initial approximations are
4 and 5.
8. Find the real root correct to three decimal places for the following equations:
pi
(i) x3 – x – 4 = 0 (ii) x3 – x2 – 1 = 0
3 2
(iii) x + x – 1 = 0 (iv) x3 – 3x – 5 = 0.
as
3
9. Solve x – 9x + 1 = 0 for the root between x = 2 and x = 4 by the method of Bisection. Also find the
smallest positive root.
10. (i) Find a real root of the equation x3 – 2x – 5 = 0 using Bisection method.
g
(ii) Find a positive root of the equation xex = 1 which lies between 0 and 1.
.c
11. If a root of f(x) = 0 lies in the interval (a, b), then find the minimum number of iterations required
when the permissible error is E.
w
12. Apply Bisection method to find a root of the equation x4 + 2x3 – x – 1 = 0 in the interval [0, 1].
13. Find the approximate value of the root of the equation 3 x − 1 + sin x = 0 by Bisection method.
w
(five iterations) [U.P.T.U. (MCA) 2008]

w
14. 3
Find a real root of x – 2x – 1 = 0 which lies between 1 and 2 by using Bisection method correct to
two places of decimals.
15. Find a positive root of the equation x3 + 3x – 1 = 0 by Bisection method.
Answers
1. (i) x: –4 –3 –2 –1 0 1 2 3 4
f(x): 1.0625 .125 – .75 – 1.5 –2 –2 –1 2 9
Roots lie in (– 3, – 2) and (2, 3).
(ii) 1.7281 in interval (1, 2).
2. 1.5121375 3. (i) (2, 3) (ii) Root lies in the interval (2, 2.25).
4. 1.4036 5. 0.517
6. (i) (– 3, – 2) (ii) Root lies in the interval (– 2.5, – 2.25) 7. 4.712389
8. (i) 1.796 (ii) 1.466 (iii) 0.755 (iv) 2.279
www.cgaspirants.com
www.cgaspirants.com
9. 2.94282, 0.111 10. (i) 2.094551482 (ii) 0.56714333
loge
FG|b − a|IJ
11. n≥
H E K 12. 0.8667605 13. 0.39188 14. 1.618 15. 0.322
loge 2
4.6 METHOD OF FALSE POSITION OR REGULA-FALSI METHOD
This is the oldest method for finding the real roots of a numerical equation.
It is sometimes known as method of linear interpolation.
In this method, we choose two points x0 and
x1 such that f(x0) and f(x1) are of opposite signs. Since B[x1, f(x1)]
the graph of y = f(x) crosses the X-axis between these
m
two points, a root must lie in between these points.
Consequently, f(x0) f(x1) < 0
o
Equation of the chord joining points {x0, f(x0)}
and {x1, f(x1)} is
.c
f ( x1 ) − f ( x0 )
y – f(x0) = ( x − x0 )
x1 − x0
The method consists in replacing the curve
AB by means of the chord AB and taking the point
n
of intersection of the chord with X-axis as an ap-
ts
ra
proximation to the root.

So the abscissa of the point where chord cuts
pi
y = 0 is given by
x2 = x0 –
LM x − x OP f (x )
1 0
…(1)
N f (x ) − f (x ) Q 0
as
1 0
which is an approximation to the root.
If now f(x0) and f(x2) are of opposite signs, then the root lies between x0 and x2.
g
So, replacing x1 by x2 in (1), we obtain the next approximation x3. However, the root could as
.c
well lie between x1 and x2 then we find x3 accordingly.

This procedure is repeated till the root is found to the desired accuracy.
w
EXAMPLES
w
Example 1. Find a real root of the equation 3x + sin x – ex = 0 by the method of false
w
position correct to four decimal places. Choose suitable initial approximations.

[U.P.T.U. 2010 , U.P.T.U. (MCA) 2009]
Sol. Let f(x) ≡ 3x + sin x – ex = 0
f(0.3) = – 0.154 i.e., (–)ve
and f(0.4) = 0.0975 i.e., (+)ve
∴ The root lies between 0.3 and 0.4.
Using Regula-Falsi method,
x1 − x0
x 2 = x0 – f ( x0 )
f ( x1 ) − f ( x0 )
(0.4) − (0.3)
= (0.3) − (− 0.154) | ∵ x0 = 0.3 and x1 = 0.4 (let)
(0.0975) − (− 0.154)
www.cgaspirants.com
www.cgaspirants.com
FG 0.1 × 0.154 IJ = 0.3612

= (0.3) +
H 0.2515 K
Now, f(x2) = f (0.3612) = 0.0019 = (+)ve
( x2 − x0 )
Now, again, x3 = x 0 – f ( x0 ) Replacing x1 by x2
f ( x2 ) − f ( x0 )
= (0.3) −
RS (0.3612) − (0.3) UV (− 0.154)
T (0.0019) − (− 0.154) W
F 0.0612 IJ (0.154) = 0.3604
= (0.3) + G
H 0.1559 K
m
Now, f(x3) = f(0.3604) = – 0.00005 = (–)ve
RS x − x UV f (x )
o
2 3
Now, again, x4 = x3 − Replacing x0 by x3
T f (x ) − f (x ) W 3
.c
2 3
= (0.3604) − M
L (0.3612 − 0.3604) OP (− 0.00005)
= 0.3604 + G
n N (0.0019) − (− 0.00005) Q
F .0008 IJ (0.00005) = 0.36042
0
H 0.00195 K
ts
ra
Since, x3 and x4 are approximately the same, hence, the required real root is 0.3604
correct to four decimal places.
Example 2. Find the root of the equation xex = cos x in the interval (0, 1) using Regula-
pi
Falsi method correct to four decimal places. [U.P.T.U. (MCA) 2006]

Sol. Let f(x) = xex – cos x
as
Since f(0.51) = – 0.02344 and f(0.52) = 0.00683

g
Let x0 = 0.51 and x1 = 0.52

.c
By Regula-Falsi method, first approximation to the root is

LM x − x OP f(x )
w
1 0
x2 = x0 –
N f (x ) − f (x ) Q
1 0
0
L 0.52 − 0.51 OP (– 0.02344) = 0.517744

w
= 0.51 – M
N 0.00683 − (− 0.02344) Q
w
f(x2) = – 0.000041
Hence, the root lies between x2 and x1.
x3 = x2 –
LM x − x OP f(x )
1 2
| Replacing x by x
N f (x ) − f (x ) Q
1 2
2 0 2
= 0.517744 – M
L 0.52 − 0.517744 OP (– 0.000041) = 0.517757
N 0.00683 − (− 0.000041) Q
Since x2 and x3 are same up to four decimal places hence the approximate real root is
0.5177.
www.cgaspirants.com
www.cgaspirants.com
Example 3. Find the root of the equation tan x + tanh x = 0 which lies in the interval
(1.6, 3.0) correct to four significant digits using method of false position.
[G.B.T.U. 2013, M.T.U. (MCA) 2012]
Sol. Let f(x) ≡ tan x + tanh x = 0
Since, f(2.35) = – 0.03 and f(2.37) = 0.009
Hence, the root lies between 2.35 and 2.37. Let x0 = 2.35 and x1 = 2.37
Using Regula-Falsi method, first approximation to the root is
x2 = x0 −
RS x − x UV f ( x )
1 0
T f (x ) − f (x ) W
1 0
0
F 2.37 − 2.35 IJ (− 0.03) = 2.36538
m
= 2.35 − G
H 0.009 + 0.03K
o
Now, f(x2) = .000719 (+)ve
.c
x3 = x0 –
n 2 0
T f (x ) − f (x ) W
2 0
ts0 | Replacing x1 by x2
F 2.36538 − 2.35 IJ (− .03)

ra
= 2.35 − G
H 0.000719 + .03 K
= 2.36502
pi
Since x2 and x3 are same up to four significant digits, hence the required root is 2.365.
Example 4. Using the method of false position, find the root of equation x6 – x4 – x3 – 1 = 0
as
upto four decimal places.

Sol. Let f(x) = x6 – x4 – x3 – 1
g
Since, f(1.4) = – 0.056 and f(1.41) = 0.102

.c
Hence, the root lies between 1.4 and 1.41. Let x0 = 1.4 and x1 = 1.41.
Using method of false position, first approximation to the root is
w
1 0
w
x2 = x0 –
T f (x ) − f (x ) W
1 0
0
F 1.41 − 1.4 IJ (− 0.056) = 1.4035

w
= 1.4 − G
H 0.102 + 0.056 K
Now, f(x2) = – 0.0016 (–)ve
x 3 = x2 −
1 2
Replacing x0 by x2
T f (x ) − f (x ) W
1 2
2
= 1.4035 − G
F 1.41 − 1.4035 IJ (− 0.0016)
H 0.102 + 0.0016 K
www.cgaspirants.com
www.cgaspirants.com
FG 0.0065 IJ (0.0016) = 1.4036

= 1.4035 +
H 0.1036 K
Now, f(x3) = – 0.00003 (–)ve
Again, the third approximation to the root is
x4 = x3 –
1 3
| Replacing x2 by x3
T f (x ) − f (x ) W
1 3
3
= 1.4036 + G
F 1.41 − 1.4036 IJ (0.00003)
H 0.102 + 0.00003 K
m
= 1.4036 + G
F 0.0064 IJ (0.00003) = 1.4036
H 0.10203 K
o
.c
Since, x3 and x4 are approximately the same up to four places of decimal, hence the
required root of the given equation is 1.4036. It is clear that –1 is also a root of the given
equation.
ts
Example 5. Find a real root of the equation x log10 x = 1.2 by Regula-Falsi method
correct to four decimal places.
n
ra
Sol. Let f(x) = x log10 x – 1.2 ...(1)
Since, f(2.74) = – .0005634 and f(2.741) = .0003087
pi
Hence, the root lies between 2.74 and 2.741. Let x0 = 2.74 and x1 = 2.741.
Using method of False position, first approximation to the root is
as
x2 = x0 −
1 0
T f (x ) − f (x ) W
1 0
0
g
R 2.741 − 2.74 UV (− .0005634)

.c
= 2.74 − S
T.0003087 − (− .0005634) W
w
= 2.74 + GH
F .001 IJ (.0005634) = 2.740646027
.0008721K
w
Now, f(x2) = – .00000006016 i.e., (–)ve

w

x3 = x2 −
1 2
| Replacing x by x2
T f (x ) − f (x ) W
1 2
2 0
= 2.740646027 – G
F 2.741 − 2.740646027 IJ (− .00000006016)
H .0003087 + .00000006016 K
= 2.740646096
Since, x2 and x3 agree up to seven decimal places, the required root correct to four decimal
places is 2.7406.
www.cgaspirants.com
www.cgaspirants.com
Example 6. (i) Apply False position method to find smallest positive root of the equation
x – e–x = 0, correct to three decimal places.
(ii) Using Regula-Falsi method, compute the smallest positive root of the equation xex – 2 = 0,
correct up to four decimal places. (U.P.T.U. 2006)
Sol. (i) Let f(x) = x – e–x
Since f(.56) = – .01121 and f(.58) = .0201
Hence, root lies between .56 and .58. Let x0 = .56 and x1 = .58
x2 = x0 −
1 0
T f (x ) − f (x ) W 0
m
1 0
= .56 − G
F .58 − .56 IJ (− .01121) = .56716
H .0201 + .01121K
o
Now, f(x2) = .00002619 i.e., (+)ve
.c
Hence the root lies between x0 and x2.
x3 = x0 −
n
2 0
ts
| Replacing x by x2
T f (x ) − f (x ) W
2 0
0 1
ra
= .56 − G
F .56716 − .56 IJ (− .01121) = .567143
H .00002619 + .01121K
pi
Since, x2 and x3 agree up to four decimal places, the required root correct to three decimal
places is 0.567.
as
(ii) Let f(x) = xex – 2

Since, f(.852) = – .00263 and f(.853) = .001715
g
.c

w
1 0
x2 = x0 −
T f (x ) − f (x ) W
1 0
0
R .853 − .852 UV (− .00263) = .852605293

w
= .852 − S
T.001715 − (− .00263) W
w
Now f(x2) = – .00000090833

Hence, root lies between x2 and x1.
x 3 = x2 −
1 2
| Replacing x by x
T f (x ) − f (x ) W
1 2
2 0 2
= (.852605293) – S
R . 853 − . 852605293 UV (− .00000090833)
T.001715 − (−.00000090833) W
= 0.852605501
Since, x2 and x3 agree upto 6 decimal places, hence the required root correct to 4 decimal
places is 0.8526.
www.cgaspirants.com
www.cgaspirants.com
Example 7. (i) Solve x3 – 5x + 3 = 0 by using Regula-Falsi method.

(ii) Use the method of false position to solve x3 – x – 4 = 0.
Sol. (i) Let f(x) = x3 – 5x + 3
Since, f(.65) = .024625 and f(.66) = – .012504
x2 = x0 –
RS x − x UV f(x )
1 0
T f (x ) − f (x ) W
1 0
0
= .65 –
FG .66 − .65 IJ (.024625) = .656632282
H − .012504 − .024625 K
m
Now, f(x2) = – .00004392
o
.c
x3 = x0 –
2 0
ts
Replacing x1 by x2
T f (x ) − f (x ) W
2 0
0
FG .656632282 − .65 IJ (.024625)

n
= .65 −
H − .00004392 − .024625 K
ra
= .656620474
Since, x2 and x3 agree up to four decimal places hence the required root is .6566 correct
pi
up to four decimal places. Similarly the other roots of this equation are 1.8342 and – 2.4909.
(ii) Let f(x) = x3 – x – 4
as
Since, f(1.79) = – .054661 and f(1.80) = .032

Hence, root lies between 1.79 and 1.80. Let x0 = 1.79 and x1 = 1.80.
g
.c
1 0
x2 = x 0 −
T f (x ) − f (x ) W
1 0
0
w
= 1.79 − S
R 1.80 − 1.79 UV (− .054661) = 1.796307
w
T.032 − (− .054661) W
w
Now, f(x2) = – .00012936

x3 = x 2 −
1 2
| Replacing x by x2
T f (x ) − f (x ) W
1 2
2 0
= 1.796307 – S
R 1.8 − 1.796307 UV (− .00012936)
T.032 − (− .00012936) W
= 1.796321.
Since, x2 and x3 are same up to four decimal places hence the required root is 1.7963
correct up to four decimal places.
www.cgaspirants.com
www.cgaspirants.com
4.7 CONVERGENCE OF REGULA-FALSI METHOD (M.T.U. 2013)
If < xn > be the sequence of approximations obtained from

( x n − xn − 1 )
x n + 1 = xn – f ( xn ) ...(1)
f ( xn ) − f ( xn − 1 )
and α be the exact value of the root of equation f(x) = 0, then
Let xn = α + en and xn + 1 = α + en + 1
where en, en + 1 being errors involved in nth and (n + 1)th approximations respectively.
Clearly, f(α) = 0. Hence, (1) gives,
( en − en − 1 )
α + en + 1 = α + en – ⋅ f (α + en )
m
f (α + en ) − f (α + en − 1 )
en − 1 f (α + en ) − en f (α + en − 1 )
o
or en + 1 =
f (α + en ) − f (α + en − 1 )
.c
LM en 2 OP LM en2 − 1 OP
=
en − 1 f (α) + en f ′(α) +
MN
LM f (α) + e f ′(α) +
en 2
2!
n
OP
PQ
LM ts
f ″ (α) + ...... − en f (α) + en − 1 f ′(α) +
MN
f ″ (α) + ...... − f (α) + en − 1 f ′(α) +
en2 − 1
2!
f ″ (α) + ......
f ″ (α) + ......
OP
PQ
MN n
PQ MN PQ
ra
2! 2!
en − 1 e n
pi
(en − 1 − en ) f (α) + (en − en − 1 ) f ″ (α) + ......

2!
=
( en − en − 1 ) ( en + en − 1 )
as
(en − en − 1 ) f ′(α) + f ″ (α) + ......

2!
en − 1 en
g
f ″ (α) + ......
2
.c
=
f ′ (α) +
FG
en + en − 1 IJ
f ″ (α) + ......
|∵ f(α) = 0
H
2 K
w
en en − 1 f ″ (α)
or en + 1 ≈ ...(2) (neglecting high powers of en, en – 1)
w
2! f ′ (α)
or en+1 = c en en – 1 ...(3)
w
1 f ″ (α)
where c= . Relation of the form (3) is called the error equation.
2 f ′ (α)
If the function f(x) in the equation f(x) = 0 is convex in the interval (x0, x1) that contains
the root, then one of the points x0 or x1 is always fixed and the other point varies with n. If the
point x0 is fixed then the function f(x) is approximated by the straight line passing through the
points (x0, f0) and (xn, fn), k = 1, 2,... . The error equation (3) becomes
en+1 = c e0 en
where e0 = x0 – α is independent of n. Therefore, we can write
en+1 = c* en
where c* = c e0 is the asymptotic error constant. Hence, the Regula-Falsi method has
linear rate of convergence.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. (i) Solve cos x = 3x – 1 correct to three decimal places using the method of False position.
(G.B.T.U. 2011)
(ii) Solve x3 – 9x + 1 = 0 for the root lying between 2 and 4 by the method of False position.
2. (i) Find real cube root of 18 by Regula-Falsi method.
(ii) Find the root of the following equation in the interval [0, 1] by Regula-Falsi method:
2x(1 – x2 + x) ln x = x2 – 1 [U.P.T.U. 2014, M.T.U. (MCA) 2010]
3. (i) Find the smallest positive root correct to three decimal places of cosh x cos x = – 1.
(ii) Use Regula-Falsi method to find the smallest positive root of the following equation correct
m
to four significant digits: x3 – 5x + 1 = 0. (M.T.U. 2012)
4. Find all the roots of cos x – x2 – x = 0 up to 5 decimal places.
5. Find the real root of the equations by using method of false position.
o
(i) x4 – x3 – 2x2 – 6x – 4 = 0 (ii) x6 – x4 – x3 – 3 = 0
.c
(iii) xex =3 (iv) x2 – log10 x – 12 = 0
6. Solve the following equations by Regula-Falsi method:
(i) (5 – x) ex = 5 near x = 5
ts
(ii) x3 + x – 1 = 0 near x = 1
(iii) 2x – log10 x = 7 lying between 3.5 and 4 (iv) x3 + x2 – 3x – 3 = 0 lying between 1 and 2
(v) x3 – 3x + 4 = 0 between – 2 and – 3
n
(vi) x4 + x3 – 7x2 – x + 5 = 0 lying between 2 and 3.
ra
7. Find the real root of the equations by using method of False position.
(i) x3 – 4x + 1 = 0 (ii) x3 – x2 – 2 = 0
pi
(iii) x3 +x–3=0 (iv) x4 – x – 10 = 0 (G.B.T.U. 2012)

8. (i) Explain Regula-Falsi method by stating at least one advantage over Bisection method.
as
(ii) Discuss method of False position. [G.B.T.U. 2011; U.P.T.U. 2009]

(iii) Illustrate False position method by plotting the function on graph and discuss the speed of
convergence to the root.
g
(iv) Find the rate of convergence for Regula-Falsi method. (M.T.U. 2013)
.c
(v) Determine the order of convergence of the iterative method

x0 f ( xk ) − xk f ( x0 )
w
xk+1 =
f ( xk ) − f ( x0 )
for finding a simple root of the equation f(x) = 0.
w
Answers
w
1. (i) 0.607 (ii) 2.942821 2. (i) 2.620741394 (ii) 1, 0.3279

3. (i) 1.875 (ii) 0.2016 4. – 1.25115 and 0.55000
5. (i) 2.7320506 (ii) 1.501844 (iii) 1.04991 (iv) 3.5425
6. (i) 4.9651142 (ii) .682327803 (iii) 3.7892782 (iv) 1.73205
(v) – 2.195823345 (vi) 2.0608526
7. (i) 1.860, .2541 (ii) 1.69562 (iii) 1.2134 (iv) 1.855
8. (v) linear : ek+1 = ce0ek.
www.cgaspirants.com
www.cgaspirants.com
4.8 NEWTON-RAPHSON METHOD (M.T.U. 2013, U.P.T.U. 2009, 2015)
This method is generally used to improve the result obtained by one of the previous methods.
Let x0 be an approximate root of f(x) = 0 and let x1 = x0 + h be the correct root so that f(x1) = 0.
Expanding f(x0 + h) by Taylor’s series, we get
h2
f(x0) + hf ′(x0) +f″(x0) + ...... = 0
2!
Since, h is small, neglecting h2 and higher powers of h, we get
f ( x0 )
f(x0) + hf ′(x0) = 0 or h=– ...(1)
f ′ ( x0 )
m
A better approximation than x0 is therefore given by x1, where
f ( x0 )
x1 = x 0 –
o
f ′ ( x0 )
Successive approximations are given by x2, x3, ....... , xn + 1, where
.c
xn + 1 = xn –
which is the Newton-Raphson formula.

n f ( xn )
f ′ ( xn )
ts ...(2) (n = 0, 1, .......)
ra
Remarks. (1) This method is useful in cases of large values of f ′(x) i.e., when the graph of f(x) while
crossing the x-axis is nearly vertical.
(2) If f ′(x) is zero or nearly 0, the method fails.
pi
(3) Newton’s formula converges provided the initial approximation x0 is chosen sufficiently close
to the root. This method is also called the method of tangents.
as
(4) This method is also used to obtain complex roots.

g
4.9 CONVERGENCE
.c
Comparing equation (2) with xn + 1 = φ (xn) of the iteration method, we get

w
f ( xn )
φ(xn) = xn + 1 = xn –
f ′( xn )
w
f ( x) f ( x) f ″( x)
In general, φ(x) = x – which gives φ′(x) =
w
f ′( x ) [ f ′( x)]2
Since, iteration method converges if | φ′ (x) | < 1
∴ Newton’s method converges if
| f (x) f ″ (x) | < [ f ′ (x)]2
in the interval considered.
Assuming f(x), f ′(x) and f ″(x) to be continuous, we can select a small interval in the
vicinity of the root α in which above condition is satisfied.
The rate at which the iteration method converges if the initial approximation to the
root is sufficiently close to the desired root is called the rate of convergence.
[M.T.U. 2014, G.B.T.U. 2011, 2012, 2013]
www.cgaspirants.com
www.cgaspirants.com
4.10 ORDER OF CONVERGENCE

[M.T.U. 2012, G.B.T.U. (MCA) 2011; U.P.T.U. 2014, 2015; G.B.T.U. (M.Tech.) 2011]
Suppose, xn differs from the root α by a small quantity en so that

xn = α + en and xn + 1 = α + e n + 1
Then equation (2) becomes,
f (α + en )
en + 1 = en –
f ′ (α + en )
en 2
f (α) + en f ′ (α) + f ″ (α) + ......
2!
= en – (By Taylor’s expansion)
m
f ′ (α) + en f ″ (α) + ......
en 2
o
en f ′ (α) + f ″ (α) + .......
= en – 2 |∵ f(α) = 0
.c
f ′ (α) + en f ″ (α) + ......
=
en 2 f ″ (α)
2[ f ′ (α) + en f ″ (α)]
en 2
n
f ″ (α)
ts | Neglect higher powers of en
ra
=
2 RS f ″ (α) UV
f ′ (α) 1 + en
T f ′ (α) W
pi
f ″ (α) R
S1 + e ff ″′ ((αα)) UVW
2 −1
e n
= .
2 f ′ (α) T
n
as
e f ″ (α) R
S1 − e ff ″′ ((αα)) + .......UVW
2
n
=
2 f ′ (α) T
g
n
.c
1 f ″ (α) e R f ″ (α) U
2
en +1
or
en2
=
2 f ′ (α)
− S V + .......
n
2 T f ′ (α) W
w
f ″ (α)
≈ (Neglecting terms containing powers of en)
w
2 f ′ (α)
Hence, by definition, the order of convergence of Newton-Raphson method is 2
w
i.e., Newton-Raphson method is quadratic convergent.

This also shows that subsequent error at each step is proportional to the square of the
previous error and as such the convergence is quadratic.
Hence, if at the first iteration, we have an answer correct to one decimal place then it
should be correct up to two places at the II iteration, up to four places at III iteration.
It means that the number of correct decimal places at each iteration is almost doubled.
∴ Method converges very rapidly.
Due to its quadratic convergence, the formula (2) is also termed as second order
formula.
www.cgaspirants.com
www.cgaspirants.com
4.11 GEOMETRICAL INTERPRETATION
Let x0 be a point near the root α of equation f(x) = 0, then Y

tangent at A {x0 , f(x0)} is
y – f (x0) = f ′(x0) (x – x0)
f ( x0 ) A {x0, f(x0)}
It cuts x-axis at x1 = x0 –
)
f(x
f ′ ( x0 )
y=
which is I approximation to root α. If A1 corresponds to x1
on the curve, then tangent at A1 will cut x-axis at x2, nearer A1
to α and is therefore II approximation to root α. A2
Repeating this process, we approach the root α quite
m
x2 x1 x0 X
rapidly. Hence, the method consists in replacing the part O
of the curve between A and x-axis by the means of the
tangent to the curve at A0. a
o
.c
4.12 NEWTON’S ITERATIVE FORMULAE FOR FINDING INVERSE, SQUARE ROOT
ETC.
1
n
tion – a = 0, which can be solved by Newton’s method.
ts
1. Inverse. The reciprocal or inverse of a number ‘a’ can be considered as a root of the equa-
x
ra
1 1
Since, f(x) = – a, f ′(x) = – 2
x x
pi
∴ Newton’s formula gives,

FG 1 − aIJ
Hx K
as
n
xn + 1 = xn +
F 1I
GH x JK
g
2
n
.c
xn + 1 = xn (2 – axn)
w
2. Square root. The square root of ‘a’ can be considered as a root of equation x2 – a = 0
w
solvable by Newton’s method.

Since, f(x) = x2 – a, f ′(x) = 2x
w
xn 2 − a
xn + 1 = xn –
2 xn
1 F a I
xn + 1 =
2
GH
xn +
xn
JK
1
3. Inverse square root. Equation is –a=0
x2
Iterative formula is
1
xn + 1 = x (3 – a xn2)
2 n
www.cgaspirants.com
www.cgaspirants.com
4. General formula for pth root. The pth root of a can be considered as a root of the
equation x p – a = 0. To solve this by Newton’s method, we have
f(x) = x p – a and hence f ′(x) = px p – 1
( xn p − a)
∴ Iterative formula is, xn + 1 = xn –
pxn p − 1
( p − 1) xn p + a
xn + 1 = [U.P.T.U. MCA (SUM) 2008]
pxn p − 1
Also, the general formula for reciprocal of pth root of a is

F p + 1 − ax I .
m
p
x n + 1 = xn GH p JK n
o
.c
4.13 ORDER OF CONVERGENCE OF NEWTON’S SQUARE ROOT FORMULA
Let a = α so that a = α2 . If we writen

xn = α
F1+ e I
GH 1 − e JK
n
n
ts
ra
=αG
F1+ e I
H 1 − e JK
n+1
then, xn + 1 ...(1)
pi
n+1
1F aI
Also, by formula, xn + 1 = Gx + J , we get
2H x K
as
n
n
1 L F1+ e I a F1− e I OP
= Mα G
2 MN H 1 − e JK α GH 1 + e JK PQ
g
n n
xn + 1 +
n n
.c
= αG
F1+ e I 2
H 1 − e JK
n
...(2) (∵ a = α2 )
w
2
n
w
Comparing (1) and (2), we get en + 1 = en2 confirming quadratic convergence of Newton’s
method.
w
4.14 ORDER OF CONVERGENCE OF NEWTON’S INVERSE FORMULA
1 1
Let α= i.e., a = . If we write xn = α(1– en)
a α
then, xn + 1 = α (1 – en + 1)
By formula, xn + 1 = xn (2 – axn), we get
xn + 1 = α(1– en) [2 – aα (1– en)] = α(1– en2) |∵ aα = 1
Comparing, we get en+1 = en2, hence, convergence is quadratic.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Using Newton-Raphson method, find the real root of the equation
3x = cos x + 1 correct to four decimal places. [G.B.T.U. (M. Tech.) 2010, 2011]
Sol. Let f(x) = 3x – cos x – 1
Since, f(0) = – 2 = (–)ve ;
f(1) = 1.4597 = (+)ve
∴ A root of f(x) = 0 lies between 0 and 1. It is nearer to 1. Let us take x0 = 0.6.
Also, f ′(x) = 3 + sin x
Newton’s iteration formula gives,
m
f ( xn )
xn + 1 = xn –
f ′ ( xn )
o
3 xn − cos xn − 1 xn sin xn + cos xn + 1
= xn – = ...(1)
.c
3 + sin xn 3 + sin xn
Put n = 0, the first approximation x1 is given by,
x1 =
3 + sin x0
n
Put n = 1 then second approximation is
=
ts
x0 sin x0 + cos x0 + 1 0.6 sin 0.6 + cos 0.6 + 1
3 + sin 0.6
= .6071
ra
x1 sin x1 + cos x1 + 1
x2 =
3 + sin x1
pi
.6071sin (.6071) + cos (.6071) + 1

= = 0.6071
3 + sin (.6071)
as
Clearly, x1 = x2. Hence, desired root is 0.6071 correct to 4 decimal places.

Example 2. Using Newton’s iterative method, find the real root of x log10 x = 1.2 correct
to six decimal places.
g
Sol. f(x) = x log10 x – 1.2

.c
∵ f(2.7) = – .0353 = (–)ve

f(2.8) = .05204 = (+)ve
w
Hence, a root of f(x) = 0 lies between 2.7 and 2.8.

w
Since | f(2.7) | < | f(2.8) |

Hence, root is nearer to 2.7.
w
Let us take x0 = 2.71

Also, f ′(x) = log10 x + log10 e = log10 x + 0.43429
Newton’s iteration formula gives,
f ( xn )
xn + 1 = xn –
f ′ ( xn )
= xn –
FG xn log10 xn − 1.2 IJ = .43429 x + 1.2
n
...(1)
H log10 xn + .43429 K log x + .43429
10 n
Put n = 0, the first approximation is
.43429 x0 + 1.2
x1 = = 2.74073 | Taking x0 = 2.71
log 10 x0 + .43429
www.cgaspirants.com
www.cgaspirants.com
Similarly, Putting n = 1, 2, in (1), we get

x2 = 2.740646
x3 = 2.740646
Since x2 and x3 are same up to 6 decimal places, hence the required root is 2.740646.
Example 3. Evaluate 12 to five decimal places by Newton’s iterative method.
Sol. Let x = 12 so that x2 – 12 = 0 ...(1)
2
Take f(x) = x – 12, Newton’s iteration formula gives,
f ( xn ) x 2 − 12 F
1 12 I
xn + 1 = xn –
f ′( xn )
= xn – n
2 xn
= GH
2
xn +
xn
JK ...(2)
m
Now since, f(3) = – 3 (–)ve
f(4) = 4 (+)ve
o
∴ The root of (1) lies between 3 and 4.
.c
Since | f(3) | < | f(4) |, hence, root is nearer to 3.
Take x0 = 3.4, equation (2) gives,
x1 =
F
GH1
2
x0 +
n I F
12
JK GH
x0
=
1
2
3.4 +
12
K
3.5
ts
IJ = 3.464706
1F 12 I
ra
x = Gx + J = 3.464102
2H x K
2 1
1
1F 12 I
pi
x = Gx + J = 3.4641016
2H x K
3 2
2
as
Since, x2 and x3 are same up to 5 decimal places,

x = 12 = 3.46410.
g
Example 4. Using Newton’s iterative method, find the real root of x sin x + cos x = 0
.c
which is near x = π correct to 3 decimal places. [G.B.T.U. (C.O.) 2010]

Sol. We have f(x) = x sin x + cos x and f ′(x) = x cos x
w
The iteration formula is,

xn sin xn + cos xn
w
x n + 1 = xn –
xn cos xn
w
x0 sin x0 + cos x0 π sin π + cos π

with x0 = π, x1 = x 0 – =π– = 2.8233
x0 cos x0 π cos π
Successive iteratives are,
x2 = 2.7986, x3 = 2.7984
Since, x2 and x3 are same up to three decimal places hence required root is 2.798.
Example 5. Find a real root of the equation x = e–x using the Newton-Raphson method.
[U.P.T.U. (MCA) 2008]
Sol. We have f(x) = x – e–x
∴ f ′(x) = 1 + e–x
www.cgaspirants.com
www.cgaspirants.com
Since, f(.5) = – .1065 (–)ve

f(.6) = .05118 (+)ve
Hence, the root lies between .5 and .6.
Since, | f(.6) | < | f(.5) | ∴ Root is nearer to 0.6.
Let us take x0 = 0.58.
Newton’s iterative formula is
f ( xn ) F
x − e − xn I
xn+1 = xn –
f ′ ( xn ) GH
= xn − n − x
1+e n
JK
Putting n = 0, we get
m
( x0 − e − x0 )
x1 = x0 – = 0.567113
(1 + e − x0 )
o
Putting n = 1, 2, we get successively
.c
x2 = 0.567143
x3 = 0.56714329
Raphson method.
n ts
Since x2 and x3 are same up to 6 decimal places, hence, the required real root is 0.567143.
Example 6. Find a positive value of (17)1/3 correct to six decimal places by Newton-
(M.T.U. 2013, U.P.T.U. 2007, 2014)
ra
Sol. Let x = (17)1/3 so that x3 – 17 = 0 ...(1)
Let x3
f(x) = – 17 ∴ f ′(x) = 3x2
pi
Since f(2.5) = – 1.375 (–)ve

and f(2.6) = 0.576 (+)ve
as

Since, | f(2.6) | < | f(2.5) |
Therefore, root is nearer to 2.6. Let us take x0 = 2.58.
g
Newton-Raphson formula is
.c
f ( xn ) F
x3 − 17 I
x n + 1 = xn – = xn – n 2 GH JK
w
f ′ ( xn ) 3 xn
Putting n = 0, 1, 2, ..., successively, we get
w
F x − 17 I = 2.571311
3
x1 = x0 – GH 3x JK
0
|∵ x0 = 2.58
w
2
0
x2 = 2.571281
x3 = 2.5712815
Since x2 and x3 are same up to 6 decimal places, hence, the required positive root is
2.571281.
Example 7. Show that the following two sequences, both have convergence of the second
order with the same limit a.
1 F
a I and, x 1 F x 2 I
xn + 1 =
2 n GH
x 1+ 2
xn JK n+1 =
2 GH
xn 3 − n .
a JK
(U.P.T.U. 2006, 2009)
www.cgaspirants.com
www.cgaspirants.com
F 1 I a
Sol. Since, xn + 1 = GH 2 nJK
x 1 + 2 , we have
xn
1 F a I – = 1 Fx a I
xn + 1 –
2 H x JK a 2 GH
a = x G1 + n
n
2 n +
xn
−2 a JK
1F a I
2
= G x −
2H x K
J = 2x1 (x –
n
n n
n a )2
1 2
Thus, en + 1 = en ...(1)
2 xn
m
which shows the quadratic convergence. Similarly for the second,
1 F
x 2 I– 1 F x 2 I
xn + 1 – a = xn 3 − nGH JK a = GH
xn 1 − n + (xn – JK a)
o
2 a 2 a
.c
=
xn
(a – xn2) + (xn – a ) = (xn –
L x ex +
a ) M1 −
N 2a
n
n a jOPQ
en + 1 =
2a
xn − a
n
2a
[2a – xn2 – xn a ] =
xn − a
2a
ts
[(a – xn2) + (a – xn a )]
F x − a I (x
ra
=– GH 2a JKn
n – a ) (xn + 2 a )
pi
( xn − a ) 2 ( xn + 2 a ) 2
en + 1 = –
(xn + 2 a ) = – . en ...(2)
2a 2a
as
which shows the quadratic convergence.

Example 8. If xn is suitable close approximation to a , show that error in the formula
F I F x I
g
2
1 a 1
GH
xn 1 + 2 JK is about 3 rd that in the formula, x = 12 x GH 3 − a JK and deduce that n
.c
xn + 1 = n+1 n
2 xn
x F 3a x I 2
w
the formula xn + 1 = n
8 HG 6+
x n
−
a JK
2
n
gives a sequence with third order convergence.
w
Sol. Since xn is very close to a

F x + 2x I e
w
en + 1 −
~ – GH 2x JKn
n
2
n
n
2 | From (2)
1
=3. e 2 ...(3)
2xn n
A simple observation shows that from (1) (see example 7) and (3), error in first formula
1
for en + 1 is about rd of that in second formula.
3
To find the rate of convergence of given formula, we have
xn F 6 + 3a − x I – 2
xn (6 xn 2 a + 3a 2 − xn 4 )
xn + 1 – a =
8 GH x a JK n
2
n
a =
8 axn 2
– a
www.cgaspirants.com
www.cgaspirants.com
6 xn 2 a + 3a 2 − xn 4 − 8 xn a a − ( xn + 3 a ) ( xn − a ) 3
= =
8 xn a 8 xn a
F x + 3 aI e
∴ en + 1 = – GH 8 x a JK
n
n
n
3
It shows that above formula has a convergence of third order.

Example 9. Show that the square root of N = AB is given by
S N
N −~ + , where S = A + B.
4 S
Sol. Let x= N ⇒ x2 – N = 0
m
Now, let f(x) = x2 – N ∴ f ′(x) = 2x
By Newton-Raphson formula,
o
f ( xn ) xn2 − N xn N
x n + 1 = xn – = xn − = +
.c
f ′ ( xn ) 2 x n 2 2 xn
A+B
then,
Let
xn + 1 =
xn =
2
A+B
4
+
n
N ~S N
A+B
− +
4 S
ts | Since S = A + B
ra
Example 10. Using the starting value 2(1 + i), solve x4 – 5x3 + 20x2 – 40x + 60 = 0 by
Newton-Raphson method given that all the roots of the given equation are complex.
pi
Sol. Let f(x) = x4 – 5x3 + 20x2 – 40x + 60

so that, f ′(x) = 4x3 – 15x2 + 40x – 40
as
∴ Newton-Raphson method gives,

f ( xn ) x 4 − 5 x 3 + 20 xn2 − 40 xn + 60
xn + 1 = xn – = xn – n 3 n
g
f ′ ( xn ) 4 xn − 15 xn2 + 40 xn − 40
.c
3 xn4 − 10 xn3 + 20 xn2 − 60

=
4 xn3 − 15 xn2 + 40 xn − 40
w
Put n = 0, take x0 = 2(1 + i) by trial, we get

w
x1 = 1.92 (1 + i)
Again, x2 = 1.915 + 1.908 i
w
Since, imaginary roots occur in conjugate pairs, roots are 1.915 ± 1.908 i up to three
places of decimal. Assuming other pair of roots to be α ± iβ, then
F α + iβ + α − iβ I
GG JJ
Sum = + 1.915 + 1.908 i = 2α + 3.83 = 5
H
+ 1.915 − 1.908 i K
⇒ α = 0.585
Also, product of roots = (α2 + β2) [(1.915)2 + (1.908)2] = 60
⇒ β = 2.805
Hence, other two roots are 0.585 ± 2.805 i.
www.cgaspirants.com
www.cgaspirants.com
Example 11. Determine the value of p and q so that rate of convergence of the iterative
N
method xn + 1 = pxn + q for computing N1/3 becomes as high as possible.
xn2
Sol. We have x3 = N
∴ f(x) = x3 – N ...(1)
Let α be the exact root, we have
α3 = N ...(2)
Substituting xn = α + en, xn + 1 = α + en + 1, N = α3 in
N
xn + 1 = pxn + q 2 , we get
m
xn
α3
o
α + en + 1 = p(α + en) + q
(α + en ) 2
.c
α3
= p(α + en) + qα 1 +
FG en IJ −2
= p (α + en ) + q
FG IJ H K
ts
2 α
en
α2 1 +
H α K
n
R| e + 3 F e I 2 U|
|T α GH α JK
= p (α + e ) + qα S1 − 2 V|
n n
ra
n − .........
W
en2
pi
= p (α + en ) + qα − 2 qen + 3q – ........
α
⇒ en + 1 = ( p + q – 1)α + ( p – 2q)en + 0(en2) + .......
as
Now, for the method to become of order as high as possible i.e., of order 2, we must have
p+q=1 and p – 2q = 0
g
2 1
.c
so that, p= and q = .
3 3
w
Example 12. The graph of y = 2 sin x and y = log x + c touch each other in the nbd. of
point x = 8. Find c and the coordinates of point of contact.
w
Sol. The graphs will touch each other if values of dy/dx at their point of contact is
same.
w
dy
For y = 2 sin x, = 2 cos x
dx
dy 1
For y = log x + c =
dx x
1
∴ 2 cos x = ⇒ x cos x – .5 = 0
x
Let f(x) = x cos x – .5
∴ f ′(x) = cos x – x sin x
xn cos xn − 0.5
∴ Newton’s iterative formula is xn + 1 = xn –
cos xn − xn sin xn
www.cgaspirants.com
www.cgaspirants.com
For n = 0, x0 = 8, first app. x1 = 7.793

Second approximation, x2 = 7.789 ≈ 7.79
Now, y = 2 sin 7.79 = 1.9960
∴ Point of contact → (7.79, 1.996)
Now, y = log x + c
⇒ 1.996 = log 7.79 + c ⇒ c = – 0.054.
Example 13. How should the constant α be chosen to ensure the fastest possible conver-
αxn + xn −2 + 1
gence with the iteration formula xn + 1 = ?
α+1
m
Sol. Since, lim xn = lim xn + 1 = ξ, we have
n→ ∞ n→∞
o
F αξ + 1 + 1I
.c
G ξ J 2
ξ = GH α + 1 JK
⇒
⇒
(α + 1)ξ3 = αξ3 + ξ2 + 1
ξ3 – ξ 2 – 1 = 0
n ts
ra
ξ can be obtained by finding a root of the equation x3 – x2 – 1 = 0.

We have f(x) = x3 – x2 – 1
pi
f ′(x) = 3x2 – 2x
as
Since, f(1.45) = (–)ve and f(1.47) = (+)ve

∴ Root lies between 1.45 and 1.47. Let x0 = 1.46.
g
By Newton-Raphson method,
.c
First approximation is
f ( x0 ) F x − x − 1I = 1.465601.
3 2
w
x1 = x 0 –
f ′ ( x0 )
= x0 – GH 3x − 2x JK
0
0
2
0
0
w
Second approximation is
f ( x1 ) F x − x − 1I = 1.46557
3 2
GH 3x − 2 x JK
w
1 1
x2 = x 1 – = x1 – 2
f ′ ( x1 ) 1 1
Hence, ξ = 1.465 correct to three decimal places.
Now, we have
α xn + xn −2 + 1
xn + 1 = ...(1)
α+1
Putting xn = ξ + en and xn+1 = ξ + en + 1 in (1), we get
(α + 1)(ξ + en + 1) = α(ξ + en) +

1 1 e
+ 1 = α(ξ + en) + 2 1 + n
FG IJ −2
+1
(ξ + en ) 2 ξ ξ H K
www.cgaspirants.com
www.cgaspirants.com
which gives,
F 2 Ie
GH
(1 + α)en+1 = α −
ξ3
JK n + O(en2)
2 2
For fastest convergence, we must have α = = = 0.636.
ξ3 (1.465) 3
Example 14. Use the Newton-Raphson method to find a solution accurate to within
10–5 for the problem, e6x + 1.441 e2x – 2.079 e4x – 0.3330 = 0; – 1 ≤ x ≤ 0. Use the initial point as
x0 = – 0.5. [U.P.T.U. MCA (SUM) 2008]
Sol. f(x) = e6x + 1.441 e2x – 2.079 e4x – 0.3330
x0 = – 0.5 | Given
m
Also f ′(x) = 6e6x + 2.882e2x – 8.316e4x
By Newton-Raphson formula,
o
f ( xn)
xn+1 = xn –
.c
f ′ ( xn )
LM e 6 xn
+ 1.441 e 2 xn − 2.079 e 4 xn − 0.3330 OP
= xn − n MN 6e 6 xn
Putting n = 0, the first approximation is, x1 = – 0.35241839

ts
+ 2.882 e 2 xn
− 8.316 e 4 xn
PQ ...(1)
Putting n = 1, the second approximation is, x2 = – 0.284980263

ra
Putting n = 2, the third approximation is, x3 = – 0.246813527

Putting n = 3, the fourth approximation is, x4 = – 0.223232852
pi
Putting n = 4, the fifth approximation is, x5 = – 0.207509457

Putting n = 5, the sixth approximation is, x6 = – 0.195400823
as
Similarly, putting n = 6, 7, 8, 9, ...... in (1), we get at last,

x12 = – 0.169602323 and x13 = – 0.16960655
g
Since x12 and x13 are same up to five decimal places, hence, the root of the given equa-
.c
tion is – 0.16960 correct up to five decimal places.

w
ASSIGNMENT
w
1. (i) By using Newton-Raphsons method, find the root of x4 – x – 10 = 0 which is near to x = 2

w
correct to three places of decimal. (U.P.T.U. 2014, G.B.T.U. 2012)

(ii) Find the root of 2 sin x – 2x + 1 = 0 correct to five significant digits with initial approximation
x0 = 1.0. (M.T.U. 2012)
2. (i) Compute one positive root of 2x – log10 x = 7 by Newton-Raphson method correct to four
decimal places. [G.B.T.U. (C.O.) 2011]
(ii) Find the real root of the equation log10 x – x + 3 = 0 correct to four demical places using
Newton-Raphson method. (G.B.T.U. 2011)
(iii) Use Newton-Raphson method to find the root of 3x – log10 x = 6 correct to four decimal places.
(G.B.T.U. 2012)
3. (i) Use the Newton-Raphson method to find a root of the equation x3 – 2x – 5 = 0.
(ii) Use Newton-Raphson method to find a root of the equation x3 – 3x – 5 = 0.
www.cgaspirants.com
www.cgaspirants.com
4. Find the real root of the equations:

(i) log x = cos x (ii) x2 + 4 sin x = 0 (iii) x3 – 4x + 1 = 0
(M.T.U. 2012)
by Newton-Raphson method correct to three decimal places.
5. Use Newton-Raphson method to obtain a root, correct to three decimal places of following
equations :
(i) sin x = 1 – x (ii) x3 – 5x + 3 = 0 (iii) x4 + x2 – 80 = 0
3 2
(iv) x + 3x – 3 = 0 (v) 4(x – sin x) = 1 (vi) x – cos x = 0
x
(vii) sin x = (viii) x log10 x = 4.77 (G.B.T.U. 2011)
2
6. (i) Explain the method of Newton-Raphson for computing roots.
m
[U.P.T.U. 2009, U.P.T.U. MCA (SUM) 2009, G.B.T.U. (MCA) 2010]
(ii) Explain the limitations of Newton-Raphson method for finding out the root of an equation.
o
(U.P.T.U. 2009)
.c
(iii) Explain the order of convergence and prove that Newton-Raphson method is second order
convergent. (M.T.U. 2012, G.B.T.U. 2010, 2011)
ts
7. If an approximate root of the equation x (1 – loge x) = 0.5 lies between 0.1 and 0.2, find the value
of the root correct to three decimal places by Newton-Raphson method. (U.P.T.U. 2015)
8. Find all the roots of cos x – x2 – x = 0 to five decimal places by Newton-Raphson method.
n
(U.P.T.U. 2006)
z
ra
x 2
9. Find all positive roots of the equation 10 e− x dt – 1 = 0 with six correct decimals. Use
0
pi
Newton-Raphson method. (U.P.T.U. 2008)

10. Use Newton-Raphson method to find the smallest positive root of the equation tan x = x.
LMHint: x lies in FG π, 3π IJ OP
as
N H 2 KQ (G.B.T.U. 2011, U.P.T.U. 2007)
x2 x3 0.3x
g
11. Find the positive root of the equation ex = 1 + x + + e correct to 6 decimal places.
2 6
.c
12. Show that the equation f(x) = cos

RS π( x + 1) UV + 0.148x – 0.9062 = 0 has one root in the interval
T 8 W
w
(– 1, 0) and one in (0, 1). Calculate the negative root correct to 4 decimals. (U.P.T.U. 2009)
w
13. (i) Use Newton’s formula to prove that square root of N can be obtained by the recursion formula,
F xi2 – N I
w
GH
xi + 1 = xi 1 −
2N JK
Hence, find the square root of
(a) 13 (U.P.T.U. 2008) (b) 21 [U.P.T.U. (MCA) 2009] (c) 35.
correct to 4 decimal places.
(ii) Find the positive value of FG 1 IJ 1/3

correct up to 4 decimal places using Newton-Raphson
H 17 K
method. (U.P.T.U. 2009)
14. Find the iterative method based on Newton-Raphson method to find cube root of N where N is a
positive real number. Apply the method for N = 18 and obtain result correct to two decimal
places. [G.B.T.U. MCA (SUM) 2010]
www.cgaspirants.com
www.cgaspirants.com
1 1
15. (i) The equation 2e–x = + has two roots greater than – 1. Calculate these roots correct
x+2 x+1
to five decimal places. (M.T.U. 2012, G.B.T.U. 2013)
(ii) Use Newton-Raphson method to find a root of non-linear equation
f (x) = x3 + 2x2 + 10x – 20 = 0 up to 10 iterations. (U.P.T.U. 2009)
16. Determine p, q and r so that the order of the iterative method
qa ra2
xn + 1 = pxn + +
xn2 xn5
for a1/3 becomes as high as possible. [U.P.T.U. (MCA) 2007]
[Hint: p + q + r = 1, p – 2q – 5r = 0, 3q + 15r = 0.]
m
Answers
1. (i) 1.855 (ii) 1.4973
o
2. (i) 3.7892 (ii) 3.5502 (iii) 2.1079
.c
3. (i) 2.094568 (ii) 2.279
4. (i) 1.303 (ii) – 1.934 (iii) 1.860, 0.2541, – 2.1147
5.
7.
(i)
(v)
0.186
0.511
1.171
(ii)
(vi)
0.657
.739
n
8. 0.55000 ; – 1.25115
ts
(iii) 2.908
(vii) 1.896
(iv) – 2.533
(viii) 6.083
ra
9. Roots lie in (0, 1) and (1, 2) ; 0.100336, 1.679631 10. 4.4934 11. 2.363376
12. – 0.5081. 13. (i) (a) 3.6055 (b) 4.5825 (c) 5.9160 (ii) 0.3889
14. 2.62 15. (i) Roots lie in (– 0.8, 0) and (0, 1) ; – 0.689752, 0.770091 (ii) 1.3688
pi
5 5 1
16. p= , q = , r = – ; Third order.
9 9 9
as
4.15 DEFINITIONS
g
.c
1. A number α is a solution of f(x) = 0 if f(α) = 0. Such a solution α is a root or a zero of f(x) = 0.

Geometrically, a root of the equation f(x) = 0 is the value of x at which the graph of y = f(x)
w
intersects x-axis.
2. If we can write f(x) = 0 as
w
f (x) = (x – α)m g(x) = 0

where g(x) is bounded and g(α) ≠ 0 then α is called a multiple root of multiplicity m. In this
w
case,
f(α) = f ′(α) = .......... = f (m – 1) (α) = 0, f (m) (α) ≠ 0
For m = 1, the number α is said to be a simple root.
4.16 METHOD FOR MULTIPLE ROOTS
If α is a multiple root of multiplicity m of the eqn. f(x) = 0 then, we have

f(α) = f ′(α) = ........ = f (m – 1)(α) = 0 and f (m)(α) ≠ 0
It can easily be verified that all the iteration methods discussed so far have only linear
rate of convergence when m > 1.
www.cgaspirants.com
www.cgaspirants.com
For example, in Newton-Raphson method, we have

ekm (m) ekm + 1 ( m + 1) ekm + 2
f(xk) = f(α + ek) = f (α ) + f (α ) + f (m + 2) (α) + .......
m! (m + 1) ! (m + 2) !
e km − 1 em (m + 1)
f ′(xk) = f ′ (α + e k ) = f ( m ) (α ) + k f (α ) + .......
( m − 1) ! m!
The error equation for N–R method becomes,
FG
ek + 1 = 1 −
1 IJ 1 f (m + 1) (α) 2
ek + O (ek3 )
H m K
ek + 2
m (m + 1) f ( m)
(α)
If m ≠ 1, we obtain,
m
FG
ek + 1 = 1 −
1 IJ
ek + O (ek2 ) ...(1)
H K
o
m
which shows that the method has only linear rate of convergence.
.c
However, if the multiplicity of the root is known in advance, we can modify the methods
convergence.
n
For example, consider Newton-Raphson method in the form
ts
by introducing parameters dependent on the multiplicity of the root to increase their order of
fk
ra
xk + 1 = x k − β ...(2)
fk ′
where β is an arbitrary parameter to be determined.
pi
If α is a multiple root of multiplicity m, we obtain from (2), the error equation
FG IJ f (m + 1) (α) 2
as
β β
ek + 1 = 1 − ek + O (ek3 )
H m K
ek + 2
m (m + 1) f ( m)
(α)
g
If the method (2) is to have the quadratic rate of convergence then the coefficient of ek
must vanish which gives
.c
β
1− = 0 or β = m
w
m
Thus, the method
w
fk
xk + 1 = xk − m
fk′
w
has quadratic rate of convergence for determining a multiple root of multiplicity m.

If the multiplicity of the root is not known in advance, then we use the following
procedure.
It is known that if f(x) = 0 has a root α of multiplicity m, then f ′(x) = 0 has the same root
α of multiplicity m – 1.
f ( x)
Hence, g(x) = has a simple root α, we can now use Newton-Raphson method
f ′ ( x)
g ( xk )
xk + 1 = x k −
g ′ ( xk )
to find the approximate value of the multiple root α.
www.cgaspirants.com
www.cgaspirants.com
Simplifying, we have
fk f k ′
xk + 1 = xk –
fk′ 2 − fk fk″
which has quadratic rate of convergence for multiple roots.
Note. If initial approximation x0 is sufficiently close to the root, then the expressions,
f ( x0 ) f ′ ( x0 ) f ″ ( x0 )
x0 − m , x0 − (m − 1) , x0 − (m − 2) will have same value.
f ′ ( x0 ) f ″ ( x0 ) f ″′ ( x0 )
EXAMPLES
m
2f(xn )
Example 1. Show that the modified Newton-Raphson method xn + 1 = xn – gives a
o
f ′ (xn )
quadratic convergence when f(x) = 0, has a pair of double roots in neighbourhood of x = xn.
.c
2f (a + en )
Sol. en + 1 = en – , where a, en and en + 1 have their usual meanings. Expanding
f ′ (a + en ) n ts
in powers of en and using f(a) = 0, f ′(a) = 0 since, x = a is a double root near x = xn, we get
LM e f ″ (a) + .......OP
2
n
2
N2! Q
ra
e n + 1 = en –
LMe f ″ (a) +
en2 OP
f ″′ (a) + .......
N Q
pi
n
2!
LM 1 f ″ (a) + 1 f ″′ (a) + .......OP

as
2 en2
= en –
N2 ! 3! Q
L
e M f ″ (a) +
e O
f ″′ (a) + ......P
g
n
Nn
2! Q
.c
L1
2e M f ″ (a) +
1 O
f ″′ (a)P
w
~ en –
−
N2 !
n
3! Q
en
w
f ″ (a) + f ″′ (a)
2!
w
en + 1 − 1 2 . f ′′′(a)
~ en
6 LM f ″ (a) + e f ′′′(a)OP
n
N 2! Q
1 2 f ′′′(a)
∴ en + 1 ≈ en
6 f ″ (a)
⇒ en + 1 ∝ en2
and hence convergence is quadratic.
Example 2. Find the double root of the equation x3 – x2 – x + 1 = 0.
Sol. Let f(x) = x3 – x2 – x + 1
so that f ′(x) = 3x2 – 2x – 1, f ″(x) = 6x – 2
www.cgaspirants.com
www.cgaspirants.com
Starting with x0 = 0.9, we have

f ( x0 ) 2 × .019
x0 – 2 = .9 – = 1.003
f ′ ( x0 ) (− .37)
f ′ ( x0 ) (−.37)
and x0 – (2 – 1) = .9 – = 1.009
f ″ ( x0 ) 3.4
The closeness of these values implies that there is a double root near x = 1.
Choosing x1 = 1.01 for next approximation, we get
f ( x1 ) 0.0002
x1 – 2 = 1.01 – 2 × = 1.0001
0.0403
m
f ′ ( x1 )
f ′ ( x1 ) .0403
and x1 – (2 – 1) = 1.01 – = 1.0001
o
f ″ ( x1 ) 4.06
.c
This shows that there is a double root at x = 1.0001 which is quite near the actual root
x = 1.
4.17 NEARLY EQUAL ROOTS

n ts
So far, Newton’s method is applicable when f ′(x) ≠ 0 in the neighbourhood of actual root x = a,
ra
i.e., in the interval (a – h, a + h).
If the quantity h is very very small, it will not satisfy the above restriction. The application
of Newton’s method will not be practical in that case. This condition arrives when roots are
pi
very close to each other.

We know that in case of double root x = a, f(x) and f ′(x) both vanish at x = a. Thus, while
as
applying Newton’s method, if xi is simultaneously near zeros of f(x) and f ′(x) i.e., f(xi) and
f ′(xi) both are very small, then it is usually practical to depart from the standard sequence
and proceed to obtain two new starting values for the two nearly equal roots.
g
To obtain these values, we first apply Newton’s method to the equation f ′(x) = 0 i.e., we
.c
use the iteration formula
f ′ ( xi )
w
x i + 1 = xi – ...(1)
f ″ ( xi )
w
with last available iterate as the initial value x0 for (1).

Suppose x = c is the solution obtained by (1).
w
Now by Taylor’s series, we have

1
f(x) = f(c) + (x – c) f ′(c) + (x – c)2 f ″(c) + .......
2
1
= f(c) + (x – c)2 f ″(c) + R |∵ f ′(c) = 0
2
Assuming R to be small, we conclude that the zeros of f(x) near x = c are approximately
given by
1 − 2f (c)
f(c) + (x – c)2 f ″(c) = 0 ⇒ x=c± ...(2)
2 f ″ (c)
www.cgaspirants.com
www.cgaspirants.com
Using these values as starting values, we can use the original iteration formula to get
two close roots of f(x) = 0.
Example. Use synthetic division to solve f(x) ≡ x3 – x2 – 1.0001 x + 0.9999 = 0 in the
neighbourhood of x = 1.
Sol. To find f(1) and f ′(1),
1 –1 –1.0001 0.9999 1
1 0 – 1.0001
1 0 – 1.0001 – 0.0002 = f (1)
1 1
m
1 1 – .0001 = f ′(1)
1
o
1
1 2= f ″(1)
.c
2
From the above synthetic division, we observe that f(1) and f ′(1) are small. Hence, there
For this, we require f ″(1).

n ts
exists two nearly equal roots. Taking x0 = 1, we will use xi + 1 = xi –
f ′ ( xi )
f ″ ( xi )
to modify the root.
ra
From the above synthetic division, we have
1
f ″(1) = 2 ⇒ f ″(1) = 4
2
pi
f ′ (1) (− .0001)
∴ First approximation x1 = 1 – =1– = 1.000025
as
f ″ (1) 4
Now, we again calculate f (x1) and f ″(x1) by synthetic division.
g
1 –1 – 1.000100 0.999900 1.000025

.c
1.000025 0.000025 – 1.000075

1 0.000025 – 1.000075 – 0.000175 = f(x1)
w
1.000025 1.000050
w
1 1.000050 – 0.000025 = f ′(x1)

1.000025
w
1
1 2.000075 = f ″(x1)
2
∴ f(1.000025) = – 0.000175 and f ″(1.000025) = 4.000150
Now, for nearly equal roots,
− 2 f (c)
x=c± , where c = 1.000025
f ″ (c)
− 2 (− .000175)
= 1.000025 ± = 1.009378, 0.990671
4.000150
www.cgaspirants.com
www.cgaspirants.com
4.18 COMPARISON OF NEWTON’S METHOD WITH REGULA-FALSI METHOD
Regula-Falsi is surely convergent while Newton’s method is conditionally convergent. But

once, Newton’s method converges, it converges faster.
In Falsi method, we calculate only one more value of function at each step i.e., f(x(n))
while in Newton’s method, we require two calculations f(xn) and f ′(xn) at each step.
∴ Newton’s method generally requires less number of iterations but requires more
time for computation at each iteration.
When f ′(x) is large near the root, correction to be applied is smaller in case of Newton’s
method and then this method is preferred while if f ′(x) is small near the root, correction to be
applied is large and curve becomes parallel to x-axis. In this case Regula-Falsi method should
m
be applied.
Newton’s method has the fastest rate of convergence. This method is quite sensitive to
o
starting value. It may diverge if f ′(x) ≈ 0 during iterative cycle.
.c
ASSIGNMENT
1.
be applied to find all the roots?
n ts
The equation f(x) = (x – 1)2 (x – 3)2 has roots at x = 1 and x = 3. Which of the following methods can
(i) Bisection method (ii) False position method (iii) Newton-Raphson method
ra
Justify your answer. (G.B.T.U. 2010)

2. A sphere of wood, 2 m in diameter, floating in water sinks to a depth d given by
pi
d3 – 3d2 + 2.5 = 0
Find d correct to two decimal places.
as
Answers
g
1. (iii) Newton-Raphson method since it deals with multiple roots as well.

.c
2. 1.16
w
4.19 INTERPOLATION [G.B.T.U. (MCA) 2010, G.B.T.U. MCA (SUM) 2010]

w
According to Theile, ‘Interpolation is the art of reading between the lines of the table’.
It also means insertion or filling up intermediate terms of the series.
w
Suppose, we are given the following values of y = f(x) for a set of values of x:
x: x0 x1 x2 ...... xn
y: y0 y1 y2 ...... yn
Thus, the process of finding the value of y corresponding to any value of x = xi between
x0 and xn is called interpolation.
Hence, interpolation is the technique of estimating the value of a function for any
intermediate value of the independent variable while the process of computing the value of
the function outside the given range is called extrapolation.
www.cgaspirants.com
www.cgaspirants.com
4.20 ASSUMPTIONS FOR INTERPOLATION
1. There are no sudden jumps or falls in the values during the period under consideration.
2. The rise and fall in the values should be uniform.
e.g., if we are given data regarding deaths in various years in a particular town and
some of the observations are for the years in which epidemic or war overtook the town then
interpolation methods are not applicable in such cases.
3. When we apply calculus of finite differences, we assume that the given set of observa-
tions are capable of being expressed in a polynomial form.
If the function f(x) is known explicitly, the value of y corresponding to any value of x can
easily be found.
If the function f(x) is not known, it is required to find a simpler function, say φ(x) such
m
that f(x) and φ(x) agree at the set of tabulated points. Such a process is called interpolation. If
φ(x) is a polynomial, then the process is called polynomial interpolation and φ(x) is called the
o
interpolating polynomial.
.c
4.21 ERRORS IN POLYNOMIAL INTERPOLATION
ts
Let the function y(x) defined by (n + 1) points (xi, yi) i = 0, 1, 2, ......, n be continuous and
differentiable (n + 1) times and let y(x) be approximated by a polynomial φn(x) of degree not
exceeding n such that
n
ra
φn(xi) = yi ; i = 0, 1, 2, ....., n ...(1)

Now problem lies in finding the accuracy of this approximation if we use φn(x) to obtain
pi
approximate values of y(x) at some points other than those defined above.
Since the expression y(x) – φn(x) vanishes for x = x0, x1, ......, xn, we put
as
y(x) – φn(x) = L Πn + 1 (x) ...(2)

where Πn+1(x) = (x – x0) (x – x1) ...... (x – xn) ...(3)
g
and L is to be determined such that equation (2) holds for any intermediate value of x say x′
where x0 < x′ < xn.
.c
y( x ′) − φ n ( x ′)
Clearly, L= ...(4)
w
Π n + 1 ( x ′)
Construct a function, F(x) = y(x) – φn(x) – L Πn + 1(x) ...(5)
w
where L is given by (4).

w
It is clear that, F(x0) = F(x1) = ...... = F(xn) = F(x′) = 0

i.e., F(x) vanishes (n + 2) times in interval [x0, xn] consequently, by repeated application of
Rolle’s theorem, F′(x) must vanish (n + 1) times, F″(x) must vanish n times in the interval
[x0, xn].
Particularly, F(n+1) (x) must vanish once in [x0, xn].
Let this point be x = ξ ; x 0 < ξ < xn .
Differentiating (5) (n + 1) times w.r.t. x and put x = ξ, we get
dn + 1
0 = (y)(n + 1) (ξ) – L (n + 1) ! ( x n + 1 ) = (n + 1) !
dx n + 1
y(n + 1) (ξ)
so that, L= ...(6)
(n + 1) !
www.cgaspirants.com
www.cgaspirants.com
Comparison of (4) and (6) gives

y(n + 1) (ξ)
y(x′) – φn(x′) = Πn + 1(x′)
(n + 1) !
Hence, required expression for error is
Π n + 1 ( x)
y(x) – φn(x) = yn + 1(ξ), x0 < ξ < xn ...(7)
(n + 1) !
Since y(x) is generally unknown, hence we do not have any information concerning
y(n + 1) (x), equation (7) is useless in practical computations.
Particularly, we will use it to determine errors in Newton’s interpolating formulae.
The various methods of interpolation are as follows:
m
(1) The method of graph (2) The method of curve fitting
(3) Use of calculus of finite difference formulae.
o
The merits of last method over others are
.c
(i) It does not assume the form of function to be known.
(ii) It is less approximate than method of graphs.
given data.
n ts
(iii) Calculations remain simple even if some additional observations are included in the
The demerit is that there is no definite way to verify whether the assumptions for the
ra
application of finite difference calculus are valid for the given set of observations.
4.22 FINITE DIFFERENCES

pi
The calculus of finite differences deals with the changes that take place in the value of the
as
function (dependent variable) due to finite changes in the independent variable.

Suppose, we are given a set of values (xi, yi) ; i = 1, 2, 3, ......, n of any function y = f(x).
g
A value of independent variable x is called argument and the corresponding value of depend-
.c
ent variable y is called entry.

Suppose that the function y = f(x) is tabulated for the equally spaced values x = x0,
w
x0 + h, x0 + 2h, ....., x0 + nh giving y = y0, y1, y2, ......, yn. To determine the values of f(x) or f ′(x)
for some intermediate values of x, given three types of differences are useful:
w
1. Forward differences. The differences y1 – y0, y2 – y1, y3 – y2, ......, yn – yn – 1 when denoted
by Δy0, Δy1, Δy2, ......, Δyn–1 respectively, are called the first forward differences where Δ is the
w
forward difference operator. (U.P.T.U. 2009)

Thus, the first forward differences are
Δyr = yr+1 – yr
Similarly, the second forward differences are defined by
Δ2yr = Δyr+1 – Δyr
Particularly, Δ2y0 = Δy1 – Δy0 = y2 – y1 – (y1 – y0) = y2 – 2y1 + y0
Similarly, Δ3y0 = y3 – 3y2 + 3y1 – y0
Δ4y0 = y4 – 4y3 + 6y2 – 4y1 + y0.
www.cgaspirants.com
www.cgaspirants.com
Clearly, any higher order difference can easily be expressed in terms of ordinates
since the coefficients occurring on R.H.S. are the binomial coefficients*. In general,
Δpyr = Δp – 1yr + 1 – Δp – 1yr defines the pth forward differences.
Following table shows how the forward differences of all orders can be formed.
Forward difference table
x y Δy Δ2 y Δ3 y Δ4 y Δ5 y
x0 y0
Δy0
x1 y1 Δ2 y 0
m
(= x0 + h) Δy1 Δ3 y 0
x2 y2 Δ2 y 1 Δ4 y 0
o
(= x0 + 2h) Δy2 Δ3 y 1 Δ5 y 0
.c
x3 y3 Δ2 y 2 Δ4 y 1
= (x0 + 3h) Δy3 Δ3 y 2
x4
= (x0 + 4h)
y4
n
Δy4
Δ2 y 3
ts
x5 y5
ra
= (x0 + 5h)
pi
Here, the first entry y0 is called leading term and Δy0, Δ2y0, ...... are called leading
differences.
as
Remark. Δ obeys distributive, commutative and index laws :
1. Δ [f(x) ± φ(x)] = Δf(x) ± Δφ (x) 2. Δ [c f (x)] = c Δ f (x) ; c is constant

g
3. Δm Δn f (x) = Δm + n f (x), m, n being (+)ve integers.

But, Δ[ f (x) . φ(x)] ≠ f (x) . Δ φ(x).
.c
2. Backward differences. The differences y1 – y0, y2 – y1, ......, yn – yn–1 when denoted by ∇y1,
w
∇y2, ......, ∇yn respectively are called first backward differences where ∇ is the backward
difference operator. (U.P.T.U. 2009)
w
Similarly, we define higher order backward differences as,

∇yr = yr – yr – 1
w
∇2yr = ∇yr – ∇yr – 1

∇3yr = ∇2yr – ∇2yr – 1 etc.
Particularly, ∇2y2 = ∇y2 – ∇y1 = y2 – y1 – (y1 – y0) = y2 – 2y1 + y0
∇3y3 = ∇2y3 – ∇2y2 = y3 – 3y2 + 3y1 – y0 etc.
∗ Δn(y0) = yn – nC1 yn – 1 + nC2yn – 2 + ...... + (– 1)n y0
www.cgaspirants.com
www.cgaspirants.com
Backward difference table
x y ∇y ∇ 2y ∇ 3y ∇ 4y ∇ 5y
x0 y0
∇y1
x1 y1 ∇ 2y 2
(= x0 + h) ∇y2 ∇ 3y 3
x2 y2 ∇ 2y 3 ∇ 4y 4
(= x0 + 2h) ∇y3 ∇ 3y 4 ∇ 5y 5
x3 y3 ∇ 2y 4 ∇ 4y 5
m
(= x0 + 3h) ∇y4 ∇ 3y 5
x4 y4 ∇ 2y 5
o
(= x0 + 4h) ∇y5
x5 y5
.c
(= x0 + 5h)
ts
3. Central differences. The central difference operator δ is defined by the relations
n
y1 – y0 = δy1/2, y2 – y1 = δy3/2, ......, yn – yn–1 = δy
n–
1.
2
ra
[G.B.T.U. 2012, U.P.T.U. 2009]

Similarly, high order central differences are defined as
pi
δy3/2 – δy1/2 = δ2y1, δy5/2 – δy3/2 = δ2y2 and so on.

These differences are shown as follows:
as
Central difference table

g
x y δy δ2y δ3y δ4y δ 5y

.c
x0 y0
δy1/2
w
x1 y1 δ2y1
δy3/2 δ3y3/2
w
x2 y2 δ2y 2 δ4y2
δy5/2 δ3y 5/2 δ5y5/2
w
x3 y3 δ2y 3 δ4y 3
δy7/2 δ3y7/2
x4 y4 δ2y4
δy9/2
x5 y5
Note 1. The central differences on the same horizontal line have same suffix.
2. It is only the notation which changes, not the differences e.g.,
y1 – y0 = Δy0 = ∇y1= δy1/2.
www.cgaspirants.com
www.cgaspirants.com
4.23 OTHER DIFFERENCE OPERATORS
1. Shift operator E. [G.B.T.U. 2012, U.P.T.U. 2009]

It is the operation of increasing the argument x by h so that
Ef(x) = f(x + h)
E2f(x) = f(x + 2h) and so on.
The inverse operator E–1 is defined by
E–1f(x) = f(x – h).
Also Enyx = yx + nh.
2. Averaging operator μ. [G.B.T.U. 2012, U.P.T.U. 2009]
It is defined by
LM y
m
μyx =
1
+y OP
2 N x+
1
2
h x−
1
2
h
Q
o
In difference calculus, E is fundamental operator and ∇, Δ, δ, μ can be expressed in
terms of E.
.c
4.24 RELATION BETWEEN OPERATORS
1. Δ=E–1
n
or E = 1 + Δ.
ts
ra
[M.T.U. (MCA) 2012, U.P.T.U. (MCA) 2009]

Proof. We know that,
pi
Δyx = yx + h – yx = Eyx – yx = (E – 1)yx

⇒ Δ=E–1
or E=1+Δ
as
2. ∇ = 1 – E–1
g
Proof. ∇yx = yx – yx – h = yx – E–1yx

.c
∴ ∇ = 1 – E–1
w
3. δ = E1/2 – E–1/2 (U.P.T.U. 2014)
– yx − h = E1/2 yx – E–1/2 yx = (E1/2 – E–1/2) yx

w
Proof. δyx = y h
x+ 2
2
w
∴ δ = E1/2 – E–1/2
1
4. μ= (E1/2 + E–1/2)
2
1 1
Proof. μyx = ( y h + yx − h ) = (E1/2 + E–1/2) yx
2 x+ 2 2
2
1
⇒ μ= (E1/2 + E–1/2)
2
5. Δ = E∇ = ∇E = δE1/2
[G.B.T.U. 2011, U.P.T.U. (MCA) 2009, M.T.U. 2012]
www.cgaspirants.com
www.cgaspirants.com
Proof. E(∇yx) = E(yx – yx–h) = yx + h – yx = Δyx

⇒ E∇ = Δ
∇(E yx) = ∇ yx + h = yx + h – yx = Δyx
⇒ ∇E = Δ
δE1/2 yx = δ y h = yx + h – yx = Δyx
x+
2
⇒ δE1/2 = Δ
6. E = ehD [U.P.T.U. (MCA) 2009]

Proof. Ef(x) = f(x + h)
m
h2
= f(x) + h f ′(x) + f ″(x) + ..... (By Taylor series)
2!
o
h2 2
.c
= f(x) + hDf(x) + D f(x) + ......
2!
LM OP
∴
= 1 + hD +
N
E = ehD
n
or
(h D) 2
2!
Δ = ehD – 1.
ts
+ ...... f(x) = ehD f(x)
Q
ra
4.25 DIFFERENCES OF A POLYNOMIAL

pi
The nth differences of a polynomial of nth degree are constant and all higher order differences
are zero when the values of the independent variable are at equal interval.
as
Let f(x) = axn + bxn – 1 + cxn – 2 + ...... + kx + l

∴ Δf(x) = f(x + h) – f(x)
g
= a[(x + h)n – xn] + b [(x + h)n – 1 – xn – 1] + ...... + kh

.c
= anhxn – 1 + b′xn – 2 + c′xn – 3 + ...... + k′x + l′ ...(1)

where b′, c′, ......, l′ are new constant coefficients.
w
∴ First differences of a polynomial of nth degree is a polynomial of degree (n – 1).

w
Similarly, Δ2f(x) = Δf(x + h) – Δf(x)

= anh [(x + h)n – 1 – xn – 1] + b′[(x + h)n – 2 – xn – 2] + ...... + k′h
w
= an(n – 1) h2xn – 2 + b″xn – 3 + ...... + k″ ...(2)

∴ Second differences represent a polynomial of degree (n – 2).
Continuing this process, for nth differences, we get a polynomial of degree zero i.e.,
Δn f(x) = an(n – 1) (n – 2) ...... 1 hn = a n ! hn
which is a constant. Hence, the (n + 1)th and higher differences of a polynomial of nth degree
will be zero. Converse of this theorem is also true.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Construct the forward difference table, given that
x: 5 10 15 20 25 30
y: 9962 9848 9659 9397 9063 8660
and point out the values of Δ2y10 , Δ4y5.
Sol. Forward difference table is as follows:
x y Δy Δ2 y Δ3 y Δ4 y
5 9962
m
– 114
10 9848 – 75
– 189 2
o
15 9659 – 73 –1
.c
– 262 1
20 9397 – 72 2
– 334 3
25
30
9063
8660
n– 403 ts
– 69
ra
From the table, Δ2y10 = – 73 and Δ4y5 = – 1.

Example 2. Construct a backward difference table for y = log x given that
pi
x: 10 20 30 40 50
y: 1 1.3010 1.4771 1.6021 1.6990
as
and find values of ∇3 log 40 and ∇4 log 50.

g
Sol. Backward difference table is:

.c
x y ∇y ∇ 2y ∇ 3y ∇ 4y
w
10 1
0.3010
w
20 1.3010 – 0.1249
0.1761 0.0738
w
30 1.4771 – 0.0511 – 0.0508

0.1250 0.0230
40 1.6021 – 0.0281
0.0969
50 1.6990
From the table, ∇3 log 40 = 0.0738 and ∇4 log 50 = – 0.0508.

Example 3. (i) Find f(6) given f(0) = – 3, f(1) = 6, f(2) = 8, f(3) = 12 ; third difference being
constant.
(ii) Find Δ10(1 – ax)(1 – bx2)(1 – cx3)(1 – dx4).
www.cgaspirants.com
www.cgaspirants.com
Sol. (i) The difference table is:
x f(x) Δf(x) Δ2f(x) Δ3f(x)

0 –3
9
1 6 –7
2 9
2 8 2
4
3 12
f(0 + 6) = E6f(0) = (1 + Δ)6f(0) = (1 + 6Δ + 15Δ2 + 20Δ3) f(0)
m
= – 3 + 6 (9) + 15 (– 7) + 20 (9) = 126.
(ii) Maximum power of x in polynomial will be 10 and co-efficient of x10 will be abcd.
o
Here, k = abcd, h = 1, n = 10
.c
∴ Expression = k hn n ! = abcd 10 !
Example 4. Evaluate:
(i) Δ tan–1 x
n
where h is the interval of differencing.
(ii) Δ2 cos 2x ts
ra
Sol. (i) Δ tan–1 x = tan–1 (x + h) – tan–1 x
= tan–1
RS x + h − x UV = tan FG h IJ
–1
pi
T 1 + x (x + h) W H 1 + hx + x K 2
(ii) Δ2 cos 2x = Δ[cos 2(x + h) – cos 2x]

as
= [cos 2(x + 2h) – cos 2(x + h)] – [cos 2(x + h) – cos 2x]
= – 2 sin (2x + 3h) sin h + 2 sin (2x + h) sin h
g
= – 2 sin h [2 cos (2x + 2h) sin h] = – 4 sin2 h cos 2(x + h).

.c
Example 5. If f(x) = exp(ax), evaluate Δnf(x).

Sol. Δeax = ea(x + h) – eax = (eah – 1)eax
w
Δ e = Δ(Δeax) = Δ[(eah – 1)eax] = (eah – 1)(eah – 1)eax = (eah – 1)2 eax

2 ax
w
Similarly, Δ3 eax = (eah – 1)3 eax

w
Δn eax = (eah – 1)n eax.
FG 1 IJ = (– 1) n ! hn
Example 6. With usual notations, prove that, Δn
H xK n .
x (x + h) ...... (x + nh)
.
FG 1IJ = Δ FG 1IJ = Δ LM 1 − 1 OP = (– 1) Δ LMΔ FG 1 − 1 IJ OP

Sol. Δn H xK n–1
H xK
Δ n–1
N x + h xQ
n–2
N H x x + hK Q
= (– 1) Δ n–2
LMFG 1 − 1IJ − FG 1 − 1 IJ OP
NH x + h x K H x + 2h x + h K Q
www.cgaspirants.com
www.cgaspirants.com
= (– 1) Δn – 2
LM 2 − 1 − 1 OP = (– 1) Δ n–2
LM − 2h OP 2
N x + h x x + 2h Q N x(x + h)(x + 2h) Q

= (– 1)2 Δn – 2
LM 2 ! h OP = (– 1) Δ
2
3 n–3
LM 3! h 3 OP
N x(x + h)(x + 2h) Q N x(x + h)(x + 2h)(x + 3h) Q
n ! hn
= (– 1)n .
x( x + h) ...... ( x + nh)
Example 7. Evaluate: (i) Δn [sin (ax + b) ] (ii) Δn [cos (ax + b)].
Sol. (i) Δ sin (ax + b) = sin [a (x + h) + b] – sin (ax + b)
m
LM FG IJ OP
ah h ah FG ah + π IJ
= 2 sin
N H2 K
cos a x +
Q
2
+ b = 2 sin
2 H
sin ax + b +
2 K
o
L ah sin FG ax + b + ah + π IJ OP
.c
sin (ax + b) = Δ M2 sin
∴ Δ2
N 2 H 2 KQ
H 2KH 2K N
n
2 ts
F ah IJ FG 2 sin ah IJ sin LMax + b + ah + π + ah + π OP
= G 2 sin
2 Q
F ah IJ sin LMax + b + 2 FG ah + π IJ OP
2
ra
= G 2 sin
H 2K N H 2 KQ
pi
Proceeding in the same manner, we get
FG ah IJ 3
LM 3(ah + π) OP
as
Δ3 sin (ax + b) = 2 sin

H 2 K sin ax + b +
N 2 Q
g
FG IJ LM OP
.c
n
ah n(ah + π)
Δn sin (ax + b) = 2 sin
H 2 K sin ax + b +
N 2 Q
w
Similarly,
F ah IJ LM FG ah + π IJ OP .
w
n
Δn cos (ax + b) = G 2 sin cos ax + b + n
(ii)
H 2K N H 2 KQ
w
LM Δf(x) OP F1 + δ I .
2
(i) Δ log f(x) = log 1 +
N f(x) Q (ii) μ = GH 4 JK
[G.B.T.U. 2013, U.P.T.U. 2009]
Sol. (i) LHS = log f (x + h) – log f (x)
= log [ f(x) + Δ f (x)] – log f (x) |∵ Δf(x) = f (x + h) – f (x)
= log
LM f (x) + Δf (x) OP = log LM1 + Δf (x) OP = RHS
N f (x) Q N f (x) Q
www.cgaspirants.com
www.cgaspirants.com
δ2 1 1/ 2
(ii) RHS = 1+ = 1+ ( E − E − 1/2 ) 2 |∵ δ = E1/2 – E–1/2
4 4
1
= 4 + (E1/2 − E −1/2 ) 2
2
1 E1/2 + E −1/2
= (E 1/2 + E − 1/2 ) 2 = = μ = LHS
2 2
FΔ I e
2
Ee x
Example 9. Prove that ex = GH E JK x .
Δ2 e x
. (M.T.U. 2013)
FΔ I e
2
m
Sol. GH E JK x = Δ2 E–1 ex = Δ2 ex – h = e–h Δ2 ex
o
E ex
RHS = e–h . Δ2 ex . = e–h . E ex = e–h ex + h = ex.
.c
Δ2 e x
Example 10. Prove that,
(i) (E1/2 + E –1/2) (1 + Δ)1/2 = 2 + Δ
(ii) Δ =
1 2
δ +δ 1 + (δ 2/4)
n ts
[G.B.T.U. 2013, U.P.T.U. 2009, G.B.T.U. (C.O.) 2011]
(U.P.T.U. 2009)
ra
2
(iii) Δ3y2 = ∇3y5.
pi
(iv) hD = – log (1 – ∇) = sin h–1 (μδ). (M.T.U. 2013, U.P.T.U. 2006)

Sol. (i) (E1/2 + E–1/2) E1/2 =E+1=1+Δ+1=Δ+2
as
1 2 δ2 1 1 1/ 2
(ii) δ +δ 1+ = (E1/2 – E–1/2)2 + (E1/2 – E–1/2) 1+ (E − E −1/ 2 ) 2
2 4 2 4
g
1 F
E1/2 + E −1/2 I
= (E + E–1 – 2) + (E1/2 – E–1/2) GH JK
.c
2 2
w
1
=
(2E – 2) = E – 1 = Δ
2
w
(iii) Δ3y2 = (E – 1)3 y2

= (E3 – 3E2 + 3E – 1) y2 = y5 – 3y4 + 3y3 – y2
w
∇3y –1
5 = (1 – E ) y5
= (1 – 3E–1 + 3E–2 – E–3) y5 = y5 – 3y4 + 3y3 – y2
(iv) hD = log E = – log (E–1) = – log (1 – ∇) |∵ E–1 = 1 – ∇
1
Also, μ= (E1/2 + E–1/2) and δ = E1/2 – E–1/2
2
1 1
∴ μδ = (E – E–1) = (ehD – e–hD) = sinh (hD)
2 2
or hD = sinh–1 (μδ).
www.cgaspirants.com
www.cgaspirants.com
Δ ∇
(i) Δ + ∇ = − , where Δ and ∇ are forward difference and backward difference
∇ Δ
operators respectively. [G.B.T.U. 2012, G.B.T.U. (MCA) 2010, 2011; U.P.T.U. 2008, 2009]
n−1
(ii) ∑Δ y
r=0
2
r = Δyn – Δy0 (iii) Δryk = ∇ryk + r.
E−1 FG IJ
Sol. (i)
Δ ∇
− =
E−1
−
1 − E −1
=
E−1
−
E H K = E−
1
= E – E–1
FG IJ
m
∇ Δ 1 − E −1 E−1 E−1 E−1 E
H E K
o
= (E – 1) + (1 – E–1) = Δ + ∇
.c
n−1 n−1
(ii) ∑ Δ2 yr = ∑ ( Δy − Δyr ) = Δy1 – Δy0 + Δy2 – Δy1 + ...... + Δyn – Δyn – 1
ts
r +1
r=0 r=0
= Δyn – Δy0.
n
FG E − 1IJ
ra
r
∇ryk + r = (1 – E–1)ryk + r = yk + r = (E – 1)r E–ryk + r = Δryk.
(iii)
H E K
pi
Fδ I =
as
1 ΔE −1 Δ 2
(i) μδ =
2
(Δ + ∇) =
2
+
2
(ii) 1 + GH 2 JK 1 + δ 2μ 2
g
[G.B.T.U. 2011] [G.B.T.U. 2013, U.P.T.U. 2014]

.c
7 4 4
(iii) ∇2 = h2D2 – h3D3 + h D . ...... (iv) ∇ – Δ = – ∇Δ
w
12
[G.B.T.U. (C.O.) 2010; G.B.T.U. (MCA) 2010, 2011]
w
1 1/2
Sol. (i) μδ = (E + E–1/2) (E1/2 – E–1/2)
w
2
1 1
= (E – E–1) = [(E – 1) + (1 – E–1)] ...(1)
2 2
1
⇒ μδ = (Δ + ∇)
2
Also from (1),
1 1 ΔE −1 Δ
μδ = [Δ + (E – 1) E–1] = (Δ + ΔE–1) = +
2 2 2 2
Hence, the results.
www.cgaspirants.com
www.cgaspirants.com
F δ I = 1 + (E
2 1/2
− E −1/2 ) 2 F
E + E –1 − 2 I
(ii) LHS = 1 + GH 2 JK 2
= 1+
2 GH JK
1
= (E + E–1)
2
RHS = 1 + δ 2μ 2
LM RS
= 1 + (E1/2 − E −1/2 ) 2 .
1 1/2
(E + E −1/2 ) 2
UVOP 1/2
N T 4 WQ
R| F (E − E ) I U| 1/2
FE + 2I F I
m
−1 2 2 1/ 2
+ E −2 −1
= S1 + G V JK = GH E +2E JK
T| H 4 JK W|
=G
H 4
o
Hence, LHS = RHS.
.c
(iii) E = ehD and ∇ = 1 – E–1
∴ ∇2 = (1 – e–hD)2
L R
n
= M1 − S1 − hD +
(hD)
−
2
(hD)
ts
+
(hD) UO
3
− ...VP
4
2
MN T 2! 3! 4! WPQ
ra
R (hD) + (hD) − (hD) + ...UV

= ShD −
2 3 4 2
T W
pi
2! 3! 4!
L R hD − (hD) + ...UOP 2
2
as
= h D M1 − S
2 2
VWP
MN T 2 6 Q
L R hD (hD) U R hD (hD) UOP
g
= h D M1 + S
2 2
MN T 2 − 6 + ...VW − 2 ST 2 − 6 + ...VWPQ
.c
2 2
w
L F 1 1I
= h D M1 − hD + G + J (hD) − ...P
O 2
2 2
N H 4 3K Q
w
F
= h D GH 1 − hD +
7 I
h D − ...J = h D – h D +
7
w
2 2
2 2
12 K 12
2
h D – ...2 3 3 4 4
F E − 1IJ – (E – 1) = (E – 1)(E – 1)
∇ – Δ = (1 – E ) – (E – 1) = GH
E K
(iv) –1 –1
= – (E – 1) (1 – E–1) = – ∇Δ
(x + 1)(m) x (m) x (m − 1)
Example 13. Prove that if m is a (+)ve integer then = +
m! m! (m − 1) !
x( x − 1) ...... ( x − m + 1) x( x − 1) ...... ( x − m + 2)
Sol. RHS = +
m! (m − 1) !
www.cgaspirants.com
www.cgaspirants.com
x( x − 1) ( x − 2) ...... ( x − m + 2)
= [(x – m + 1) + m]
m!
( x + 1) x( x − 1)( x − 2) ...... ( x − m + 2) ( x + 1) (m)
= = = LHS.
m! m!
Example 14. Given u0 + u8 = 1.9243, u1 + u7 = 1.9590
u2 + u6 = 1.9823, u3 + u5 = 1.9956. Find u4.
Sol. Taking Δ8 u 0 = 0
⇒ (E – 1)8 u0 = 0
⇒ u8 – 8c1u7 + 8c2u6 – 8c3u5 + 8c4u4 – 8c5u3 + 8c6u2 – 8c7u1 + 8c8u0 = 0
m
⇒ (u0 + u8) – 8(u1 + u7) + 28(u2 + u6) – 56(u3 + u5) + 70 u4 = 0
⇒ u4 = 0.99996. (After putting the values)
o
ASSIGNMENT
.c
1. (i) Form a table of differences for the function:
ts
f(x) = x3 + 5x – 7 for x = –1, 0, 1, 2, 3, 4, 5. Continue the table to obtain f(6) and f(7).
(ii) If y = x3 + x2 – 2x + 1, calculate values of y for x = 0, 1, 2, 3, 4, 5 and form the difference table.
n
Find the value of y at x = 6 by extending the table and verify that the same value is obtained
by substitution.
ra
2. Prove that: D sin–1 x = sin–1 [(x + 1) 1 − x2 – x 1 − ( x + 1)2 ].

pi
3. (i) Write forward difference table for

x: 10 20 30 40
y: 1.1 2.0 4.4 7.9.
as
(ii) Assuming that the following values of y belong to a polynomial of degree 4, compute the next
three values:
g
x: 0 1 2 3 4 5 6 7
y: 1 –1 1 –1 1 — — —
.c
4. Construct the table of differences for the data below.

w
x: 0 1 2 3 4
f(x) : 1.0 1.5 2.2 3.1 4.6
w
Evaluate Δ3 f(2).
5. Prove that:
w
1
(i) ∇ = ΔE–1 = E–1Δ = 1 – E–1 (ii) E1/2 = μ +δ
2
(iii) δ = ΔE–1/2 = ∇E1/2 (iv) E = (1 – ∇)–1
(v) Δ∇ = ∇Δ = δ2 (G.B.T.U. 2011) (vi) δ = Δ(1 + Δ)–1/2 = ∇(1 – ∇)–1/2 (U.P.T.U. 2009)
1
(vii) δ2E = Δ2 (G.B.T.U. 2012) (viii) μδ = (E – E–1) (G.B.T.U. 2012)
2
(ix) Δ ≡ ∇ (1 – ∇)–1 [M.T.U. (MCA) 2012]
6. ux is a function of x for which fifth differences are constant and
u1 + u7 = –786, u2 + u6 = 686, u3 + u5 = 1088. Find u4.
7. Prove that:
(i) u4 = u3 + Δu2 + Δ2u1 + Δ3u1 (ii) u4 = u0 + 4Δu0 + 6Δ2u–1 + 10Δ3u–1.
www.cgaspirants.com
www.cgaspirants.com
8. Evaluate:
F 2 I ; h = 1.
x
(i) Δ(eax log bx) (ii) Δ GH ( x + 1)!JK
Answers
1. (i) 239, 371 (ii) 1, 1, 9, 31, 73, 141; 241
3. (ii) 31, 129, 351 4. 0.4
6. 570.9.
8.
LM
(i) eax e ah log 1 +
FG h IJ
+ ( e ah − 1) log bx
OP (ii) –
x
2x .
N H x K Q ( x + 2) !
m
4.26 FACTORIAL NOTATION
o
A product of the form x(x – 1)(x – 2) ...... (x – r + 1) is denoted by [x]r and is called a factorial.
.c
Particularly, [x] = x; [x]2 = x(x – 1); [x]3 = x (x – 1)(x – 2) etc.
In case the interval of differencing is h then
[x]n = x(x – h) (x – 2h) ...... (x – n − 1 h)
n ts
Factorial notation helps in finding the successive differences of a polynomial directly by
simple rule of differentiation.
ra
4.27 TO SHOW THAT (i) Δn[x]n = n ! (ii) Δn + 1 [x]n = 0

pi
Δ[x]n = [(x + h)]n – [x]n

= (x + h)(x + h – h) (x + h – 2h) ...... (x + h – n − 1 h)
as
– x(x – h) (x – 2h) ...... (x – n − 1 h)

g
= x(x – h) ...... (x – n − 2 h) [x + h – (x – nh + h)] = nh [x]n – 1

.c
Similarly, Δ2[x]n = Δ[nh [x]n–1] = nh Δ[x]n – 1 = n(n – 1) h2 [x]n – 2

w
Δn [x]n = n(n – 1) ..... 2 . 1 . hn – 1 (x + h – x) = n ! hn

w
Also, Δn + 1[x]n
= n ! hn – n ! hn = 0
when h = 1, Δ[x]n = n[x]n – 1 and Δn [x]n = n !
w
Hence, the result of differencing [x]r is analogous to that of differencing xr when h = 1.
1
4.28 TO SHOW THAT x(–n) = , THE INTERVAL OF DIFFERENCING
( x + n)(n)
BEING UNITY
By definition of x(n), we have

x(n) = (x – n − 1 h) x(n–1) ...(1)
when interval of differencing is h.
∴ When n = 0, we have x(0) = (x + h) x(–1) ...(2)
www.cgaspirants.com
www.cgaspirants.com
Since, Δx(n) = nhx(n – 1) ...(3)

when n = 1, Δx(1) = hx(0).
⇒ Δx = h x(0) ⇒ h = hx(0) ⇒ x(0) = 1
1
From (2), x(–1) = ...(4)
( x + h)
when n = –1, from (1),
x(–1) = (x + 2h) x(–2)
1 1
⇒ = (x + 2h) x(–2) ⇒ x(–2) =
x+h ( x + h)( x + 2h)
1
In general, x(–n) = ...(5)
m
( x + h)( x + 2h) ...... ( x + nh)
1
x(–n) =
o
( x + nh) (n)
Here x(–n) is called reciprocal factorial where n is a (+)ve integer.
.c
1
Particular case. When h = 1, x(–n) = .
4.29 MISSING TERM TECHNIQUE

n ts
( x + n) ( n)
ra
Suppose n values out of (n + 1) values of y = f(x) are given, the values of x being equidistant.
Let the unknown value be N. We construct the difference table.
Since only n values of y are known, we can assume y = f(x) to be a polynomial of degree
pi
(n – 1) in x.
Equating to zero the nth difference, we can get the value of N.
as
EXAMPLES
g
Example 1. Express f(x) = x4 – 12x3 + 24x2 – 30x + 9 and its successive differences in
.c
factorial notation. Hence show that Δ5f(x) = 0.

f(x) = A[x]4 + B[x]3 + C[x]2 + D[x] + E
w
Sol. Let
Using method of synthetic division, we divide by x, x – 1, x – 2, x – 3 etc. successively,
w
then
1 1 – 12 24 – 30 9=E
w
1 – 11 13
2 1 – 11 13 – 17 = D
2 – 18
3 1 –9 –5=C
3
4 1 –6=B
1=A
www.cgaspirants.com
www.cgaspirants.com
Hence, f(x) = [x]4 – 6[x]3 – 5[x]2 – 17[x] + 9

∴ Δf(x) = 4[x]3 – 18[x]2 – 10[x] – 17
Δ2f(x) = 12[x]2 – 36[x] – 10
Δ3f(x) = 24[x] – 36
Δ4f(x) = 24 and Δ5f(x) = 0.
Example 2. Obtain the function whose first difference is 9x2 + 11x + 5.
Sol. Let f(x) be the required function so that Δf(x) = 9x2 + 11x + 5
Let 9x2 + 11x + 5 = 9[x]2 + A[x] + B = 9x(x – 1) + Ax + B
Putting x = 0, B=5 and x = 1, A = 20
m
∴ Δf(x) = 9[x]2 + 20[x] + 5
o
Integrating, we get
.c
[ x]3 [ x] 2
f(x) = 9 + 20 + 5[x] + c
3 2
where c is the constant of integration.

n ts
= 3x(x – 1) (x – 2) + 10x(x – 1) + 5x + c = 3x3 + x2 + x + c
ra
Example 3. Find the missing values in the table:
x: 45 50 55 60 65
pi
y: 3 — 2 — – 2.4.
Sol. Difference table is as follows:
as
x y Δy Δ2 y Δ3 y
45 3
g
y1 – 3
.c
50 y1 5 – 2y1
2 – y1 3 y1 + y3 – 9
w
55 2 y1 + y3 – 4
y3 – 2 3.6 – y1 – 3y3
w
60 y3 – 0.4 – 2y3
– 2.4 – y3
w
65 – 2.4
As only three entries y0, y2, y4 are given, the function y can be represented by a second
degree polynomial.
∴ Δ 3 y0 = 0 and Δ 3y 1 = 0
⇒ 3y1 + y3 = 9 and y1 + 3y3 = 3.6
y1 = 2.925, y2 = 0.225.
www.cgaspirants.com
www.cgaspirants.com
Example 4. Find the relation between α, β and γ in order that α + βx + γx2 may be
expressible in one term in the factorial notation.
Sol. Let f(x) = α + βx + γx2 = (a + bx)(2)
where a and b are certain unknown constants.
Now, (a + bx)(2) = (a + bx) [a + b(x – 1)]
= (a + bx) (a – b + bx) = (a + bx)2 – ab – b2x
= (a2 – ab) + (2ab – b2)x + b2x2 = α + βx + γx2
Comparing the co-efficients of various powers of x, we get
α = a2 – ab, β = 2ab – b2, γ = b2
Eliminating a and b from the above equations, we get γ2 + 4αγ = β2
m
Example 5. (i) Estimate the missing term in the following table:
x: 0 1 2 3 4
o
y = f(x): 1 3 9 ? 81. [G.B.T.U. (C.O.) 2010]
.c
(ii) Find the missing term in the table:
x: 2 3 4 5 6
f(x): 45 49.2 54.1
Sol. (i) We are given Four values
∴
n
Δ4f(x) = 0 ∀ x
? 67.4
ts [M.T.U. 2013, U.P.T.U. 2008]
ra
⇒ (E – 1)4 f(x) = 0 ∀ x
⇒ (E4 – 4E3 + 6E2 – 4E + 1) f(x) = 0 ∀ x
⇒ f(x + 4) – 4f(x + 3) + 6f(x + 2) – 4f(x + 1) + f(x) = 0 ∀ x
pi
where interval of differencing is 1.

Now putting x = 0, we obtain,
as
f(4) – 4f(3) + 6f(2) – 4f(1) + f(0) = 0 ...(1)

⇒ 81 – 4f(3) + 54 – 12 + 1 = 0 (From table)
g
⇒ 4f(3) = 124 ⇒ f(3) = 31.

.c
(ii) We are given four values.

∴ Δ4 f(x) = 0 ∀ x
w
⇒ (E – 1)4 f(x) = 0 ∀ x
w
⇒ (E4 – 4E3 + 6E2 – 4E + 1) f(x) = 0 ∀x

⇒ f(x + 4) – 4f(x + 3) + 6f(x + 2) – 4f(x + 1) + f(x) = 0 ∀ x
w
where interval of differencing is 1.

Now, putting x = 2, we obtain,
f(6) – 4f(5) + 6f(4) – 4f(3) + f(2) = 0
⇒ 67.4 – 4f(5) + (6 × 54.1) – (4 × 49.2) + 45 = 0
⇒ 4f(5) = 240.2
⇒ f(5) = 60.05
Example 6. Estimate the production for 1964 and 1966 from the following data:
Year: 1961 1962 1963 1964 1965 1966 1967
Production: 200 220 260 — 350 — 430
Sol. Since five figures are known, assume all the fifth order differences as zero. Since
two figures are unknown, we need two equations to determine them.
www.cgaspirants.com
www.cgaspirants.com
Hence, Δ5y0 = 0 and Δ5y1 = 0

⇒ (E – 1)5y0 = 0 and (E – 1)5y1 = 0
⇒ y5 – 5y4 + 10y3 – 10y2 + 5y1 – y0 = 0
and y6 – 5y5 + 10y4 – 10y3 + 5y2 – y1 = 0
Substituting the known values, we get
y5 – 1750 + 10y3 – 2600 + 1100 – 200 = 0
and 430 – 5y5 + 3500 – 10y3 + 1300 – 220 = 0
⇒ y5 + 10y3 = 3450 ...(1)
and – 5y5 – 10y3 = – 5010 ...(2)
Adding (1) and (2), we get
m
– 4y5 = – 1560 ⇒ y5 = 390
From (1), 390 + 10y3 = 3450
o
⇒ 10y3 = 3060 ⇒ y3 = 306
.c
Hence production for year 1964 = 306
ts
and production for year 1966 = 390.
ASSIGNMENT
n
ra
1. (i) Estimate the missing term in the following:
x: 1 2 3 4 5 6 7
y: 2 4 8 — 32 64 128
pi
Explain why the result differs from 16?

(ii) Find the missing value of the following data:
as
x: 1 2 3 4 5
f(x) : 7 × 13 21 37
g
2. (i) From the following data, find the value of U47:

.c
U46 = 0.2884, U48 = 0.5356, U49 = 0.6513, U50 = 0.7620.

[Hint. Δ4 Ux = 0 ⇒ (E – 1)4 Ux = 0.]
w
(ii) Given: log 100 = 2, log 101 = 2.0043, log 103 = 2.0128, log 104 = 2.0170. Find log 102.
w
3. (i) Determine the missing values in the following table:

x: 0 5 10 15 20 25
w
y: 6 10 — 17 — 31 (G.B.T.U. 2011)
(ii) x: 1 1.5 2 2.5 3 3.5 4
f(x) : 6 ? 10 20 ? 15 5
[G.B.T.U. 2010; G.B.T.U. (C.O.) 2011]
(iii) x: 1 2 3 4 5 6 7 8
f(x) : 1 8 ? 64 ? 216 343 512
(iv) x: 2 2.1 2.2 2.3 2.4 2.5 2.6
y: 0.135 — 0.111 0.100 — 0.082 0.074
(v) x: 10 15 20 25 30 35
f(x) : 43 — 29 32 — 77 [M.T.U. 2012, G.B.T.U. 2011]
www.cgaspirants.com
www.cgaspirants.com
(vi) x: 1 2 3 4 5 6 7 8
f(x) : 2 4 8 — 32 — 128 256
Explain why the results differ from 16 and 64. [G.B.T.U. 2013]
x−1
4. Express f(x) = in terms of negative factorial polynomials.
( x + 1) ( x + 3)
Answers
1. 2x
(i) 16.1, is not a polynomial (ii) 9.5
2. (i) 0.4147 (ii) 2.0086
3. (i) 13.25, 22.5 (ii) 0.222, 22.022
(iii) 27, 125 (iv) 0.123, 0.0904
m
(v) 33.933, 46.733 (vi) 16.257, 63.476; 2x is not a polynomial.
4. x(–1) – 4x(–2) + 4x(–3)
o
We now proceed to study the use of finite difference calculus for the purpose of
.c
interpolation. Thus we shall do in following cases which are as follows:
(i) The value of argument in given data varies by an equal interval. The technique is
called an interpolation with equal intervals.
ts
(ii) The values of argument are not at equal intervals. This is known as interpolation
with unequal intervals.
n
ra
4.30 NEWTON’S FORMULAE FOR INTERPOLATION

Newton’s formula is used for constructing the interpolation polynomial. It makes use of divided
pi
differences. This result was first discovered by the Scottish mathematician James Gregory
(1638–1675) a contemporary of Newton.
as
Gregory and Newton did extensive work on methods of interpolation but now the for-
mula is referred to as Newton’s interpolation formula. Newton has derived general forward
and backward difference interpolation formulae.
g
.c
4.31 NEWTON’S GREGORY FORWARD INTERPOLATION FORMULA

(M.T.U. 2013)
w
Let y = f(x) be a function of x which assumes the values f(a), f(a + h), f(a + 2h), ......., f(a + nh)
w
for (n + 1) equidistant values a, a + h, a + 2h, ......, a + nh of the independent variable x. Let f(x)
be a polynomial of nth degree.
w
Let f(x) = A0 + A1 (x – a) + A2 (x – a) (x – a – h) + A3 (x – a) (x – a – h) (x – a – 2h ) +
....... + An (x – a) ...... (x – a – n − 1h) …(1)
where A0, A1, A2 , ......., An are to be determined.

Put x = a, a + h, a + 2h, ......., a + nh in (1) successively.
For x = a, f(a) = A0 ...(2)
For x = a + h, f(a + h) = A0 + A1h
⇒ f(a + h) = f(a) + A1h | By (2)
Δf (a)
⇒ A1 = ...(3)
h
www.cgaspirants.com
www.cgaspirants.com
For x = a + 2h,
f(a + 2h) = A0 + A1 (2h) + A2 (2h) h
= f(a) + 2h
RS Δf (a) UV + 2h 2 A2
T h W
⇒ 2h2A2 = f(a + 2h) – 2f(a + h) + f(a) = Δ2f(a)
Δ2 f (a)
⇒ A2 =
2 ! h2
Δ3 f (a)
Similarly, A3 = and so on.
3 ! h3
m
Δn f (a)
Thus, An = .
n ! hn
o
From (1),
Δ2 f (a)
.c
Δ f (a)
f(x) = f(a) + (x – a) + (x – a) (x – a – h) + .......
h 2 ! h2
x−a
n ts + (x – a) ...... (x – a – n − 1 h)
Δn f (a)
n ! hn
Put x = a + hu ⇒ u= , we have
ra
h
Δ f (a) (hu) (hu − h) 2
f(a + hu) = f(a) + hu + Δ f (a) +......
pi
h 2 ! h2
(hu) (hu − h) (hu − 2h) ....... (hu − n − 1 h) n
+ Δ f(a)
as
n ! hn
u(u − 1) 2
g
⇒ f(a + hu) = f(a) + uΔ f(a) + Δ f(a) + ...

2!
.c
u(u − 1)(u − 2) ... (u − n + 1) n

+ Δ f(a)
n!
w
This formula is particularly useful for interpolating the values of f(x) near the beginning
w
of the set of values given. h is called interval of differencing while Δ is forward difference
operator.
w
EXAMPLES
Example 1. The population of a town in the decennial census was as given below. Esti-
mate the population for the year 1895. (G.B.T.U. 2011)
Year x: 1891 1901 1911 1921 1931
Population y: 46 66 81 93 101
(in thousands)
Sol. Here a = 1891, h = 10,
a + hu = 1895 ⇒ 1891 + 10 u = 1895 ⇒ u = 0.4
www.cgaspirants.com
www.cgaspirants.com
The difference table is as under:
1891 46
20
1901 66 –5
15 2
1911 81 –3 –3
12 –1
m
1921 93 –4
8
o
1931 101
.c
Applying Newton’s forward difference formula,
n
y(1895) = y(1891) + u Δy(1891) +
2!
ts
u(u − 1) 2
Δ y(1891) +
u(u − 1)(u − 2) 3
3!
Δ y(1891)
ra
u(u − 1)(u − 2)(u − 3) 4

+ Δ y(1891)
4!
pi
(0.4)(0.4 − 1)
as
⇒ y(1895) = 46 + (0.4)(20) + (– 5)
2
g
(0.4)(0.4 − 1)(0.4 − 2) ( 0.4 )( 0.4 − 1)( 0.4 – 2)( 0.4 − 3)

+ ( 2) + (– 3)
.c
6 24
y(1895) = 54.8528 thousands = 54852.8 ≈ 54853

w
⇒
Hence the population for the year 1895 is 54853 approximately.
w
Example 2. From the following table of half-yearly premium for policies maturing at
w
different ages, estimate the premium for policies maturing at age of 46. (U.P.T.U. 2014)
Age 45 50 55 60 65
Premium 114.84 96.16 83.32 74.48 68.48

(in rupees)
Sol. Here h = 5, a = 45, a + hu = 46

∴ 45 + 5u = 46 ⇒ u = 0.2
www.cgaspirants.com
www.cgaspirants.com
The difference table is:
Age Premium Δy Δ2 y Δ3 y Δ4 y
(x) (in rupees)
(y)
45 114.84
– 18.68
50 96.16 5.84
– 12.84 – 1.84
55 83.32 4 0.68
– 8.84 – 1.16
m
60 74.48 2.84
–6
o
65 68.48
.c
By Newton’s forward difference formula,
y46 = y45 + u Δy45 +

u(u − 1) 2
2!
Δ y45 +
n 3! ts
u(u − 1)(u − 2) 3
Δ y45 +
u(u − 1)(u − 2)(u − 3) 4
4!
Δ y45
(0.2)(0.2 − 1)
ra
= 114.84 + (0.2)(–18.68) + (5.84)
2!
pi
(0.2)(0.2 − 1)(0.2 − 2) (0.2)(0.2 − 1)(0.2 − 2)(0.2 − 3)

+ (– 1.84) + (0.68)
3! 4!
as
= 110.525632
Hence the premium for policies maturing at the age of 46 is ` 110.52.
g
Example 3. From the following table, find the value of e0.24 :

.c
x: 0.1 0.2 0.3 0.4 0.5

w
e x: 1.10517 1.22140 1.34986 1.49182 1.64872.

Sol. The difference table is:
w
x 105y 105Δy 105Δ2y 105 Δ3y 104Δ4y

w
0.1 110517
11623
0.2 122140 1223
12846 127
0.3 134986 1350 17
14196 144
0.4 149182 1494
15690
0.5 164872
www.cgaspirants.com
www.cgaspirants.com
Here h = 0.1. ∴ 0.24 = 0.1 + (0.1) u or u = 1.4

Newton-Gregory forward difference formula is
u(u − 1) 2 u(u − 1)(u − 2) 3
y(0.24) = y(0.1) + u Δ y(0.1) + Δ y( 0.1) + Δ y(0.1)
2! 3!
u(u − 1)(u − 2)(u − 3) 4
+ Δ y(0.1)
4!
u(u − 1)
⇒ 105 y(0.24) = 105 y(0.1) + u 105 Δy(0.1) + 105 Δ2y(0.1)
2!
u(u − 1)(u − 2) u(u − 1)(u − 2)(u − 3)
+ 105 Δ3y(0.1) + 105 Δ4y(0.1)
3! 4!
m
(1.4)(1.4 − 1)
⇒ 105 y(0.24) = 110517 + (1.4)(11623) + (1223)
2
o
(1.4)(1.4 − 1)(1.4 − 2) (1.4)(1.4 − 1)(1.4 − 2)(1.4 − 3)
+ (127) + (17)
.c
3! 4!
= 127124.9088
∴ y(0.24) = e.24 = 1.271249088
ts
Example 4. The following table gives the population of a town during the last six cen-
n
suses. Estimate the population in 1913 by Newton’s forward difference formula
ra
Year 1911 1921 1931 1941 1951 1961
Population 12 15 20 27 39 52
pi
(in thousands)
[U.P.T.U. (MCA) 2009]

as
Sol. Here, a = 1911, h = 10, x = 1913

x − a 1913 − 1911
g
∴ u= = = 0.2
h 10
.c
Forward difference table is

w
x f(x) Δf(x) Δ2f(x) Δ3f(x) Δ4f(x) Δ5f(x)

w
1911 12
w
3
1921 15 2
5 0
1931 20 2 3
7 3 – 10
1941 27 5 –7
12 –4
1951 39 1
13
1961 52
www.cgaspirants.com
www.cgaspirants.com
Newton’s forward difference formula is
u(u − 1) 2 u(u − 1) (u − 2) 3
f(1913) = f(1911) + uΔf(1911) + Δ f (1911) + Δ f (1911)
2! 3!
u(u − 1)(u − 2)(u − 3) 4 u(u − 1)(u − 2)(u − 3)(u − 4) 5

+ Δ f (1911) + Δ f (1911)
4! 5!
(0.2)(− 0.8) (0.2)(− 0.8)(− 1.8)(− 2.8)

= 12 + (0.2)(3) + (2) + (3)
2 4!
m
(0.2)(− 0.8)(− 1.8)(− 2.8)(− 3.8)
+ (− 10)
5!
o
f(1913) = 12.08384 thousands
.c
Hence the population of the town in the year 1913 ~
− 12083.84 = 12084 (approximately).
Example 5. From the table, estimate the number of students who obtained marks between
40 and 45.
Marks: 30–40
n
40–50
ts50–60
(M.T.U. 2013)
60–70 70–80
ra
No. of students: 31 42 51 35 31.
Sol. Difference table is:
pi
Marks less than

(x) y Δy Δ2 y Δ3 y Δ4 y
as
40 31
g
42
.c
50 73 9
51 – 25
w
60 124 – 16 37
35 12
w
70 159 –4
31
w
80 190
We shall find y45, number of students with marks less than 45.
a = 40, h = 10, a + hu = 45.
∴ 40 + 10u = 45 ⇒ u = 0.5
u(u − 1) 2
y(45) = y(40) + u Δ y(40) + Δ y(40)
2!
u(u − 1)(u − 2) 3 u(u − 1)(u − 2)(u − 3) 4
+ Δ y(40) + Δ y(40)
3! 4!
www.cgaspirants.com
www.cgaspirants.com
(0.5)(0.5 − 1) (0.5)(0.5 − 1)(0.5 − 2)

= 31 + (0.5)(42) + ( 9) + (– 25)
2 6
(0.5)(0.5 − 1)(0.5 − 2)(0.5 − 3)
+ (37)
24
= 47.8672 ≈ 48
Hence number of students getting marks less than 45 = 48
By number of students getting marks less than 40 = 31
Hence number of students getting marks between 40 and 45 = 48 – 31 = 17.
Example 6. Find the cubic polynomial which takes the following values:
m
x: 0 1 2 3
f(x): 1 2 1 10.
o
Sol. Let us form the difference table:
.c
x y Δy Δ2 y Δ3 y
1
1
2
n
1 ts –2
–1 12
ra
2 1 10
9
pi
3 10
Here, h = 1. Hence using the formula, x = a + hu

as
and choosing a = 0, we get x = u

∴ By Newton’s forward difference formula,
g
x( x − 1) 2 x ( x − 1) ( x − 2) 3
.c
y = y0 + x Δy0 + Δ y0 + Δ y0
2! 3!
w
x( x − 1) x( x − 1) ( x − 2)
= 1 + x(1) + (– 2) + (12)
2! 3!
w
= 2x3 – 7x2 + 6x + 1
w
Hence the required cubic polynomial is

y = f(x) = 2x3 – 7x2 + 6x + 1.
Example 7. The following table gives the marks secured by100 students in the Numerical
Analysis subject:
Range of marks: 30–40 40–50 50–60 60–70 70–80
No. of students: 25 35 22 11 7
Use Newton’s forward difference interpolation formula to find.
(i) the number of students who got more than 55 marks.
(ii) the number of students who secured marks in the range from 36 to 45.
www.cgaspirants.com
www.cgaspirants.com
Sol. The given table is re-arranged as follows:

Marks obtained No. of students
Less than 40 25
Less than 50 60
Less than 60 82
Less than 70 93
Less than 80 100
(i) Here, a = 40, h = 10, a + hu = 55
∴ 40 + 10u = 55 ⇒ u = 1.5
First, we find the number of students who got less than 55 marks.
m
The difference table is as under:
Marks obtained No. of Δy Δ2 y Δ3 y Δ4 y
o
less than students = y
.c
40 25
35
50
60
60
82
n 22
11
ts
– 13
– 11
2
7
5
ra
70 93 –4
7
80 100
pi
Applying Newton’s forward difference formula,

as
u(u − 1) 2 u(u − 1)(u − 2) 3 u(u − 1)(u − 2)(u − 3) 4

y55 = y40 + uΔy40 + Δ y40 + Δ y40 + Δ y40
2! 3! 4!
g
(1.5)(.5) (1.5)(.5)(− .5) (1.5)(.5)(− .5)(− 1.5)

.c
= 25 + (1.5)(35) + (−13) + (2) + (5)

2! 3! 4!
w
= 71.6171875 ≈ 72
There are 72 students who got less than 55 marks.
w
∴ No. of students who got more than 55 marks = 100 – 72 = 28

(ii) To calculate the number of students securing marks between 36 and 45, take the
w
difference of y45 and y36.

x − a 36 − 40
u= = = – .4
h 10
45 − 40
Also, u= = .5
10
By Newton’s forward difference formula.
u(u − 1) 2 u(u − 1)(u − 2) 3 u(u − 1)(u − 2)(u − 3) 4
y36 = y40 + u Δ y40 + Δ y40 + Δ y40 + Δ y40
2! 3! 4!
www.cgaspirants.com
www.cgaspirants.com
(− 0.4)(− 1.4) (− 0.4)(− 1.4)(− 2.4)

= 25 + (– 0.4)(35) + (− 13) + (2)
2! 3!
( − 0.4 )( − 1.4 )( − 2.4 )( − 3.4 )

+ (5) = 7.864 ≈ 8
4!
u(u − 1) 2 u(u − 1)(u − 2) 3
Also, y45 = y40 + u Δ y40 + Δ y40 + Δ y40
2! 3!
u(u − 1) (u − 2) (u − 3) 4
+ Δ y40
4!
m
(0.5)( − 0.5) (0.5)( − 0.5)( − 1.5)
= 25 + (0.5)(35) + ( − 13) + (2)
2 6
o
(0.5)( − 0.5)( − 1.5)( − 2.5)
+ (5)
.c
24
= 44.0546 ≈ 44.
= y45 – y36 = 44 – 8 = 36.

n ts
Hence, the number of students who secured marks in the range from 36 to 45 is
Example 8. The following are the numbers of deaths in four successive ten year age
ra
groups. Find the number of deaths at 45–50 and 50–55.

Age group: 25–35 35–45 45–55 55–65
pi
Deaths: 13229 18139 24225 31496.

Sol. Difference table of cumulative frequencies:
as
Age up to No. of deaths Δf(x) Δ2f(x) Δ3f(x)

g
x f(x)
.c
35 13229
18139
w
45 31368 6086
24225 1185
w
55 55593 7271
31496
65 87089
w
Here, h = 10, a = 35, a + hu = 50

∴ 35 + 10u = 50 ⇒ u = 1.5
u(u − 1) 2 u(u − 1)(u − 2) 3
y50 = y35 + u Δ y35 + Δ y35 + Δ y35
2! 3!
(1.5)(0.5) (1.5)(0.5)(– 0.5)
= 13229 + (1.5)(18139) + ( 6086) + (1185)
2 6
= 42645.6875 ≈ 42646
∴ Deaths at the age between 45 – 50 is 42646 – 31368 = 11278
and Deaths at the age between 50 – 55 is 55593 – 42646 = 12947.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. Estimate the value of f(22) from the following available data:

x: 20 25 30 35 40 45
f(x): 354 332 291 260 231 204
2. Given that:
x: 1 2 3 4 5 6
y(x): 0 1 8 27 64 125
Find the value of f(2.5). [U.P.T.U. (MCA) 2008]
3. (i) The table below gives value of tan x for 0.10 ≤ x ≤ 0.30.
m
x: 0.10 0.15 0.20 0.25 0.30
tan x: 0.1003 0.1511 0.2027 0.2553 0.3093
o
Evaluate tan 0.12 using Newton’s forward difference formula.
(ii) Find the value of sin 52° from the following table using Newton’s forward difference formula:
.c
θ: 45° 50° 55° 60°
ts
sin θ: 0.7071 0.7660 0.8192 0.8660
4. (i) Fit a polynomial of degree 3 and hence determine y(3.5) for the following data:
n
x: 3 4 5 6
y: 6 24 60 120 [M.T.U. 2013, G.B.T.U. 2011, 2012]
ra
(ii) Find the interpolating polynomial to the following data and hence find the value of y for x = 5:
x: 4 6 8 10
pi
f(x): 1 3 8 16 (G.B.T.U. 2013)

(iii) Express the value of θ in terms of x using the following data:
as
x: 40 50 60 70 80 90
θ: 184 204 226 250 276 304
g
Also find θ at x = 43. (M.T.U. 2012)

5. (i) Obtain the value of f(3.5) from the following data:
.c
x: 3 4 5 6 7
w
f(x): 3 6.6 15 22 35 (G.B.T.U. 2010)

(ii) Use Newton-Gregory formula to compute y at x = 24 from the following data:
w
x: 21 25 29 33 37
y: 18.4 17.8 17.1 16.3 15.5 [G.B.T.U. (C.O.) 2011]
w
6. (i) Find the cubic polynomial which takes the following values:
y(0) = 1, y(1) = 0, y(2) = 1 and y(3) = 10
Hence or otherwise obtain y(4).
(ii) Find the polynomial interpolating the data:
x: 0 1 2
y: 0 5 2 (U.P.T.U. 2008)
7. Ordinates f(x) of a normal curve in terms of standard deviation x are given as
x: 1.00 1.02 1.04 1.06 1.08
f(x): 0.2420 0.2371 0.2323 0.2275 0.2227
Find the ordinate for standard deviation x = 1.025.
www.cgaspirants.com
www.cgaspirants.com
8. Find the number of men getting wages between ` 10 and ` 15 from following table:
Wages (in `): 0–10 10–20 20–30 30–40
Frequency: 9 30 35 42 (G.B.T.U. 2011)
9. Following are the marks obtained by 492 candidates in a certain examination
Marks: 0–40 40–45 45–50 50–55 55–60 60–65
No. of candidates: 210 43 54 74 32 79
Find out the number of candidates who secured
(a) more than 48 but not more than 50 marks
(b) less than 48 but not less than 45 marks.
10. Find the number of students from the following data who secured marks not more than 45
m
Marks range: 30–40 40–50 50–60 60–70 70–80
No. of students : 35 48 70 40 22
o
11. Use Newton’s forward difference formula to obtain the interpolating polynomial f (x) satisfying
.c
the following data:
x: 1 2 3 4
f(x): 26 18 4 1
ts
If another point x = 5, f(x) = 26 is added to the above data, will the interpolating polynomial be
n
the same as before or different. Explain why?
ra
12. Find the polynomial of degree four which takes the following values:
x: 2 4 6 8 10
y: 0 0 1 0 0 (U.P.T.U. 2007)
pi
13. Use Newton’s method to find a polynomial p(x) of lowest possible degree such that p(n) = 2n for
n = 0, 1, 2, 3, 4.
as
14. Find the order of the polynomial which might be suitable for the following function:
x: 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7
g
f(x): 0.577 0.568 0.556 0.540 0.520 0.497 0.471 0.442

.c
Also find the value of f(2.15) using difference formulae. [G.B.T.U. (MCA) 2010]
w
Answers
1. 352.22304 2. 3.375
w
3. (i) 0.1205 (ii) 0.7880

3 2 11 1 2 11
w
4. (i) x3 – 3x2 + 2x, 13.125 (ii) x − x + 6, 1.625 (iii) x + x + 124, 189.79

8 4 100 10
5. (i) 3.28125 (ii) 17.9571
6. (i) x3 – 2x2 + 1 ; 33 (ii) 9x – 4x2
7. 0.23589625 8. 15
9. (a) 27 (b) 27 10. 51
17 3 193
11. x – 20x2 + x + 11; no change since third differences are constant.
6 6
1 x4 x 3 11 2 7 x
12. (x4 – 24x3 + 196x2 – 624x + 640) 13. − + x + +1
64 24 12 12 12
14. 7th, 0.562425293.
www.cgaspirants.com
www.cgaspirants.com
4.32 NEWTON’S GREGORY BACKWARD INTERPOLATION FORMULA

[U.P.T.U. MCA (SUM) 2008]
Let y = f(x) be a function of x which assumes the values f(a), f(a + h), f (a + 2h), ......,
f(a + nh) for (n + 1) equidistant values a, a + h, a + 2h, ......, a + nh of the independent variable x.
Let f(x) be a polynomial of nth degree.
Let, f(x) = A0 + A1(x – a – nh) + A2 (x – a – nh) (x – a – n − 1 h) + ......
+ An (x – a – nh) (x – a – n − 1 h) ...... (x – a – h)
where A0, A1, A2, A3, ......, An are to be determined. ...(1)
Put x = a + nh, a + n − 1 h, ......, a in (1) respectively.
Put x = a + nh, then f (a + nh) = A0 ...(2)
m
Put x = a + (n – 1) h, then
f(a + n − 1 h) = A0 – h A1 = f(a + nh) – h A1 | By (2)
o
∇ f (a + nh)
⇒ A1 = ...(3)
.c
h
Put x = a + (n – 2)h, then
⇒
f (a + n − 2 h) = A0 – 2hA1 + (– 2h) (– h) A2
n
∇ 2 f (a + nh)
ts
2 ! h2 A2 = f(a + n − 2 h) – f(a + nh) + 2 ∇ f(a + nh) = ∇ 2 f(a + nh)
ra
A2 = ...(4)
2 ! h2
∇ n f (a + nh)
Proceeding, we get An = ...(5)
pi
n ! hn
Substituting the values in (1), we get
as
∇f (a + nh)
f(x) = f(a + nh) + (x – a – nh) + ...... + (x – a – nh) (x – a – n − 1 h)
h
g
∇ n f (a + nh)
..... (x – a – h) …(6)
n ! hn
.c
Put x = a + nh + uh, then

w
x – a – nh = uh
and x – a – (n – 1)h = (u + 1)h
w
x – a – h = (u + n − 1) h
w
∴ (6) becomes,
u(u + 1) 2
f(x) = f(a + nh) + u ∇ f(a + nh) + ∇ f(a + nh)
2!
∇ n f (a + nh)
+ ...... + uh . (u + 1)h ..... (u + n − 1)(h)
n ! hn
u(u + 1) 2
or f(a + nh + uh) = f(a + nh) + u ∇ f(a + nh) + ∇ f(a + nh)
2!
u(u + 1) ...... (u + n − 1) n
+ ...... + ∇ f(a + nh)
n!
This formula is useful when the value of f(x) is required near the end of the table.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. The population of a town was as given. Estimate the population for the
year 1925.
Years (x): 1891 1901 1911 1921 1931
Population (y): 46 66 81 93 101
(in thousands)
Sol. Here, a + nh = 1931, h = 10, a + nh + uh = 1925
1925 − 1931
∴ u= = – 0.6
10
m
Difference table is:
x y ∇y ∇ 2y ∇ 3y ∇ 4y
o
1891 46
.c
20
1901 66 –5
1911 81
n 15
12
ts–3
2
–1
–3
ra
1921 93 –4
8
1931 101
pi
Applying Newton’s Backward difference formula, we get

as
u(u + 1) 2
y1925 = y1931 + u ∇ y1931 + ∇ y1931
2!
g
u(u + 1)(u + 2) 3 u(u + 1)(u + 2)(u + 3) 4

+ ∇ y1931 + ∇ y1931
.c
3! 4!
( − 0.6)( 0.4 ) ( − 0.6)( 0.4 )(1.4 )
w
= 101 + (– 0.6)(8) + ( −4 ) + (–1)

2! 3!
w
( − 0.6)(0.4)(1.4)(2.4)
+ (–3)
4!
w
= 96.8368 thousands.
Hence the population for the year 1925 = 96836.8 ≈ 96837.
Example 2. The population of a town is as follows:
Year: 1921 1931 1941 1951 1961 1971
Population: 20 24 29 36 46 51
(in Lakhs)
Estimate the increase in population during the period 1955 to 1961.
Sol. Here, a + nh = 1971, h = 10, a + nh + uh = 1955
∴ 1971 + 10u = 1955 ⇒ u = – 1.6
www.cgaspirants.com
www.cgaspirants.com
Difference table is:
x y ∇y ∇ 2y ∇ 3y ∇ 4y ∇ 5y
1921 20
4
1931 24 1
5 1
1941 29 2 0
7 1 –9
1951 36 3 –9
10 –8
m
1961 46 –5
5
1971 51
o
.c
Applying Newton’s backward difference formula, we get
u(u + 1) 2 u(u + 1)(u + 2) 3
y1955 = y1971 + u ∇ y1971 +
+
4!
n 2!
u(u + 1)(u + 2)(u + 3) 4 ts
∇ y1971 +
∇ y1971 +
3!
∇ y1971
u(u + 1)(u + 2)(u + 3)(u + 4) 5

5!
∇ y1971
ra
(− 1.6)(− 0.6) (− 1.6)(− 0.6)(0.4)

= 51 + (– 1.6)(5) + (– 5) + (– 8)
2! 6
pi
(− 1.6)(− 0.6)(.4)(1.4) (− 1.6)(− 0.6)(0.4)(1.4)(2.4)

+ (– 9) + (– 9)
24 120
= 39.789632
as
∴ Increase in population during period 1955 to 1961 is

= 46 – 39.789632 = 6.210368 Lakhs = 621036.8 ≈ 621037.
g
Example 3. Evaluate from following table f(3.8) to three significant figures using Gregory-
.c
Newton backward interpolation formula

x: 0 1 2 3 4
w
f(x): 1 1.5 2.2 3.1 4.6 (U.P.T.U. 2009)

Sol. Here, a + nh = 4, h = 1, a + nh + uh = 3.8
w
3.8 − 4
∴ u= = – 0.2
w
1
Backward difference table is
x f(x) ∇ f(x) ∇2 f(x) ∇3f(x) ∇4f(x)

0 1
0.5
1 1.5 0.2
0.7 0
2 2.2 0.2 0.4
0.9 0.4
3 3.1 0.6
1.5
4 4.6
www.cgaspirants.com
www.cgaspirants.com
By Newton’s backward difference formula,

u (u + 1) 2 u (u + 1) (u + 2) 3
f(3.8) = f(4) + u∇f(4) + ∇ f (4) + ∇ f (4)
2 3!
u (u + 1) (u + 2) (u + 3) 4
+ ∇ f (4)
4!
( − 0.2) ( 0.8) ( − 0.2) ( 0.8) (1.8)

= 4.6 + (– 0.2) (1.5) + ( 0.6) + ( 0.4 )
2 3!
( − 0.2) ( 0.8) (1.8) ( 2.8)
m
+ ( 0.4 ) = 4.21936.
4!
Example 4. Given log x for x = 40, 45, 50, 55, 60 and 65 according to the following table:
o
x: 40 45 50 55 60 65
.c
log x: 1.60206 1.65321 1.69897 1.74036 1.77815 1.81291
Find the value of log 5875.
n ts [M.T.U. (MCA) 2012]
ra
x 105 log x = 105yx 105 ∇yx 105 ∇2yx 105 ∇3yx 105∇4 yx 105 ∇5yx
40 160206
pi
5115
45 165321 – 539
as
4576 102
50 169897 – 437 – 25
g
4139 77 5
55 174036 – 360 – 20
.c
3779 57
60 177815 – 303
w
3476
w
65 181291
Newton’s Backward difference formula is

w
u(u + 1) 2
f(a + nh + uh) = f(a + nh) + u∇f(a + nh) + ∇ f(a + nh)
2!
u(u + 1)(u + 2) 3 u(u + 1)(u + 2)(u + 3) 4
+ ∇ f(a + nh) + ∇ f(a + nh)
3! 4!
u(u + 1)(u + 2)(u + 3)(u + 4) 5
+ ∇ f(a + nh) ...(1)
5!
First we shall find the value of log(58.75).
Here, a + nh = 65, h = 5, a + nh + uh = 58.75
∴ 65 + 5u = 58.75 ⇒ u = – 1.25
www.cgaspirants.com
www.cgaspirants.com
From (1),
( −1.25)( − 0.25)
105 f(58.75) = 181291 + (–1.25)(3476) + (– 303)
2!
( − 1.25)( − 0.25)( 0.75) ( − 1.25)( − 0.25)( 0.75)(1.75)
+ (57) + (– 20)
3! 4!
( − 1.25)( − 0.25)( 0.75)(1.75)( 2.75)
+ (5)
5!
⇒ 105 f(58.75) = 176900.588
∴ f(58.75) = log 58.75 = 176900.588 × 10–5 = 1.76900588
Hence, log 5875 = 3.76900588 | ∵ Mantissa remain the same
m
Example 5. From the following table of half-yearly premium for policies maturing at
different ages, estimate the premium for policy maturing at the age of 63:
o
Age: 45 50 55 60 65
.c
Premium: 114.84 96.16 83.32 74.48 68.48
(in rupees)
Age Premium
n
∇y
ts ∇ 2y ∇ 3y ∇ 4y
ra
(x) (in rupees)

(y)
pi
45 114.84
– 18.68
as
50 96.16 5.84
– 12.84 – 1.84
55 83.32 4 0.68
g
– 8.84 – 1.16
.c
60 74.48 2.84
–6
w
65 68.48
w
Here a + nh = 65, h = 5, a + nh + uh = 63
∴ 65 + 5u = 63 ⇒ u = – 0.4
w
By Newton’s backward difference formula,

u(u + 1) 2 u(u + 1)(u + 2) 3
y(63) = y(65) + u∇y(65) + ∇ y(65) + ∇ y(65)
2! 3!
u(u + 1)(u + 2)(u + 3) 4
+ ∇ y(65)
4!
( − 0.4 )( 0.6)
= 68.48 + (– 0.4)(– 6) + ( 2.84 )
2
( − 0.4 )( 0.6)(1.6) ( − 0.4 )( 0.6)(1.6)( 2.6)
+ ( − 116
. )+ (.68)
6 24
= 70.585152
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. Find a polynomial of degree three using Newton-Gregory backward difference formula which
takes the following values. Hence find y(7):
x: 3 4 5 6
y: 6 24 60 120 (G.B.T.U. 2012)
2. The population of a town in decennial census is as under. Estimate the population for the year
1955:
Year: 1921 1931 1941 1951 1961
Population (in lacs): 46 66 81 93 101 (M.T.U. 2012)
3. Estimate the value of f (42) from the following available data:
m
x: 20 25 30 35 40 45
f(x): 354 332 291 260 231 204
o
4. The table below gives the value of tan x for 0.10 ≤ x ≤ 0.30:
.c
x: 0.10 0.15 0.20 0.25 0.30
y = tan x: 0.1003 0.1511 0.2027 0.2553 0.3093
Find:
5. (i) Given:
x:
(i) tan 0.50
1 2
(ii) tan 0.26
3
n
4 5
ts
(iii) tan 0.40.
6 7 8
ra
f(x): 1 8 27 64 125 216 343 512
Find f (7.5) using Newton’s Backward difference formula.
(ii) Compute f(8) from the following data:
pi
x: 1 3 5 7 9
f(x): 9 21 81 237 537 (G.B.T.U. 2013)
as
6. Using Newton’s backward difference formula, find the value of e–1.9 from the following table of
values of e–x:
g
x: 1 1.25 1.50 1.75 2.00

–x
e : 0.3679 0.2865 0.2231 0.1738 0.1353
.c
7. If y(10) = 35.3, y(15) = 32.4, y(20) = 29.2, y(25) = 26.1, y(30) = 23.2 and y(35) = 20.5, find y(12)
using Newton’s forward as well as backward interpolation formula. Also explain why the differ-
w
ence (if any) in the result occur. (U.P.T.U. 2007)

8. From the following table of values of x and f(x), determine (i) f(0.23) (ii) f (0.29):
w
x: 0.20 0.22 0.24 0.26 0.28 0.30

w
f(x): 1.6596 1.6698 1.6804 1.6912 1.7024 1.7139

9. From the following table, find the value of tan 17°
θ°: 0 4 8 12 16 20 24
tan θ°: 0 0.0699 0.1405 0.2126 0.2867 0.3640 0.4402
10. From the following table:
x: 10° 20° 30° 40° 50° 60° 70° 80°
cos x: 0.9848 0.9397 0.8660 0.7660 0.6428 0.5000 0.3420 0.1737
Calculate cos 25° and cos 73° using Gregory Newton formula. (U.P.T.U. 2006)
11. Use Newton–Gregory formula to interpolate the value of y at x = 36 from the following data:
x: 21 25 29 33 37
y: 18.4 17.8 17.1 16.3 15.5 (U.P.T.U. 2014)
www.cgaspirants.com
www.cgaspirants.com
Answers
1. x3 – 3x2 + 2x, 210 2. 9683680 3. 219
4. (i) 0.5543 (ii) 0.2662 (iii) 0.4241
5. (i) 421.875 (ii) 366 6. 0.1496
7. 34.22007, 34.30866 8. (i ) 1.6751 (ii) 1.7081
9. 0.3057 10. 0.9063, 0.2923 11. 15.698
4.33 INTERPOLATION BY UNEVENLY SPACED POINTS

The interpolation formulae derived so far possess the disadvantage of being applicable only
m
to equally spaced values of the argument. It is then desirable to develop interpolation formulae
for unequally spaced values of x. We shall study two such formulae:
(1) Lagrange’s interpolation formula
o
(2) Newton’s general interpolation formula with divided differences.
.c
4.34 LAGRANGE’S INTERPOLATION FORMULA [G.B.T.U. (MCA) 2010, 2011]
n
polynomial corresponding to the arguments x0, x1, x2, ......, xn.
ts
Let f(x0), f(x1) ,......, f(xn) be (n + 1) entries of a function y = f(x), where f(x) is assumed to be a
The polynomial f(x) may be written as

ra
f(x) = A0 (x – x1) (x – x2) ...... (x – x n) + A1(x – x0)(x – x2) ...... (x – xn)

+ ...... + An (x – x0) (x – x1) ...... (x – xn –1) ...(1)
pi
where A0, A1, ......, An are constants to be determined.

Putting x = x0, x1, ......, xn in (1), we get
as
f(x0) = A0 (x0 – x1) (x0 – x2) ....... (x0 – xn)

f ( x0 )
g
∴ A0 = ...(2)
( x0 − x1) ( x0 − x2 ) ...... ( x0 − xn )
.c
f(x1) = A1 (x1 – x0) (x1 – x2) ...... (x1 – xn)

f ( x1)
∴ A1 = ...(3)
w
( x1 − x0 ) ( x1 − x2 ) ...... ( x1 − xn )
w
f ( xn )
Similarly, An = ...(4)
( xn − x0 ) ( xn − x1) ...... ( xn − xn − 1)
w
Substituting the values of A0, A1, ......, An in equation (1), we get
( x − x1) ( x − x2 ) ...... ( x − xn ) ( x − x0 ) ( x − x2 ) ...... ( x − xn )

f(x) = f(x0) + f(x1)
( x0 − x1) ( x0 − x2 ) ...... ( x0 − xn ) ( x1 − x0 ) ( x1 − x2 ) ...... ( x1 − xn )
( x − x0 ) ( x − x1) ...... ( x − xn − 1)
+ ...... + f(xn)
( xn − x0 ) ( xn − x1) ...... ( xn − xn − 1)
This is called Lagrange’s interpolation formula.
www.cgaspirants.com
www.cgaspirants.com
4.35 ANOTHER FORM OF LAGRANGE’S FORMULA
Lagrange’s formula can also be put in the form

n
φ ( x) f ( x r )
Pn(x) = ∑ ( x − x ) φ′ ( x )
r=0 r r
n
φ(x) = Π ( x – x r )
LM d {φ (x)}OP
where
r=0
and φ′(xr) =
N dx Q x = xr
We have the Lagrange’s formula,

n
( x − x0 ) ( x − x1 ) ... ( x − x r − 1 ) ( x − x r + 1 ) ... ( x − xn )
∑ (x
m
Pn(x) = f ( xr )
r=0 r − x0 ) ( x r − x1 ) ... ( x r − x r − 1 ) ( x r − x r + 1 ) ... ( x r − xn )
R φ (x) U | R |UV
o
n
f ( xr )
= ∑ ST x − x VW S| ( x |W ...(1)
.c
r=0 T r r − x0 ) ( x r − x1 ) ... ( x r − x r − 1 ) ( x r − x r + 1 ) ... ( x r − xn )
ts
n
Now, φ(x) = Π ( x – x r )
r=0
n
= (x – x0)(x – x1) ..... (x – xr – 1) (x – xr) (x – xr + 1) ..... (x – xn)
∴ φ′(x) = (x – x1) (x – x2) ..... (x – xr) ..... (x – xn) + (x – x0) (x – x2) ..... (x – xr) ..... (x – xn)
ra
+ ..... + (x – x0) (x – x1) ..... (x – xr – 1) (x – xr + 1) ..... (x – xn) + .....

+ (x – x0) (x – x1) ..... (x – xr) ..... (x – xn – 1)
pi
⇒ φ′(xr) = [φ′ ( x)] x = xr

= (xr – x0) (xr – x1) ..... (xr – xr – 1) (xr – xr + 1) ..... (xr – xn) ...(2)
as
n
φ ( x) f ( x r )
Hence from (1), Pn(x) = ∑ ( x − x ) φ′ ( x ) |Using (2)
g
r=0 r r
.c
EXAMPLES
w
Example 1. Using Lagrange’s interpolation formula, find y(10) from the following table:
w
x 5 6 9 11
w
y 12 13 14 16 (U.P.T.U. 2009)
Sol. Here, x0 = 5, x1 = 6, x2 = 9, x3 = 11
f(x0) = 12, f(x1) = 13, f(x2) = 14, f(x3) = 16

Lagrange’s formula is
( x − x1 ) ( x − x2 ) ( x − x3 ) ( x − x 0 ) ( x − x2 ) ( x − x3 )
f(x) = f ( x0 ) + f ( x1 )
( x0 − x1 ) ( x0 − x2 ) ( x0 − x3 ) ( x1 − x0 ) ( x1 − x2 ) ( x1 − x3 )
( x − x0 ) ( x − x1 ) ( x − x3 ) ( x − x0 ) ( x − x1 ) ( x − x2 )
+ f ( x2 ) + f ( x3 )
( x2 − x0 ) ( x2 − x1 ) ( x2 − x3 ) ( x3 − x0 ) ( x3 − x1 ) ( x3 − x2 )
www.cgaspirants.com
www.cgaspirants.com
( x − 6) ( x − 9) ( x − 11) ( x − 5) ( x − 9) ( x − 11)
f(x) = (12) + (13)
(5 − 6) (5 − 9) (5 − 11) (6 − 5) (6 − 9) (6 − 11)
( x − 5) ( x − 6) ( x − 11) ( x − 5) ( x − 6) ( x − 9)
+ (14) + (16)
(9 − 5) (9 − 6) (9 − 11) (11 − 5) (11 − 6) (11 − 9)
1 13 7
= – ( x − 6) ( x − 9) ( x − 11) + ( x − 5) ( x − 9) ( x − 11) – ( x − 5) ( x − 6) ( x − 11)
2 15 12
4
+ ( x − 5) ( x − 6) ( x − 9)
15
Putting x = 10, we get
1 13
f(10) = – (10 − 6) (10 − 9) (10 − 11) + (10 − 5) (10 − 9) (10 − 11)
2 15
m
7 4
– (10 − 5) (10 − 6) (10 − 11) + (10 − 5) (10 − 6) (10 − 9)
12 15
o
= 14.66666667
.c
Hence, y(10) = 14.66666667.
Example 2. Compute the value of f(x) for x = 2.5 from the following table:
x:
f(x): 1
1 2
8
using Lagrange’s interpolation method.
3
27
n 4
64 ts
ra
Sol. Here x0 = 1, x1 = 2, x2 = 3, x3 = 4
f(x0) = 1, f(x1) = 8, f(x2) = 27, f(x3) = 64
pi
( x − x1 ) ( x − x2 ) ( x − x3 ) ( x − x0 ) ( x − x 2 ) ( x − x3 )
f(x) = f ( x0 ) + f ( x1 )
as
( x0 − x1 ) ( x0 − x2 ) ( x0 − x3 ) ( x1 − x0 ) ( x1 − x2 ) ( x1 − x3 )
( x − x0 ) ( x − x1 ) ( x − x3 ) ( x − x0 ) ( x − x1 ) ( x − x2 )
g
+ f ( x2 ) + f ( x3 )
( x2 − x0 ) ( x2 − x1 ) ( x2 − x3 ) ( x3 − x0 ) ( x3 − x1 ) ( x3 − x2 )
.c
( x − 2) ( x − 3) ( x − 4) ( x − 1) ( x − 3) ( x − 4) ( x − 1) ( x − 2) ( x − 4)
= (1) + (8) + (27)
w
(1 − 2) (1 − 3) (1 − 4) (2 − 1) (2 − 3) (2 − 4) (3 − 1) (3 − 2) (3 − 4)
( x − 1) ( x − 2) ( x − 3)
w
+ (64)
(4 − 1) (4 − 2) (4 − 3)
w
Putting x = 2.5, we get

1
f(2.5) = − (2.5 – 2) (2.5 – 3) (2.5 – 4) + 4(2.5 – 1) (2.5 – 3) (2.5 – 4)
6
27 32
– (2.5 – 1) (2.5 – 2) (2.5 – 4) + (2.5 – 1) (2.5 – 2) (2.5 – 3)
2 3
= 15.625.
Example 3. Find the cubic Lagrange’s interpolating polynomial from the following data:
x: 0 1 2 5
f(x): 2 3 12 147.
Sol. Here x0 = 0, x1 = 1, x2 = 2, x3 = 5
f(x0) = 2, f(x1) = 3, f(x2) = 12, f(x3) = 147
www.cgaspirants.com
www.cgaspirants.com
( x − x1 ) ( x − x2 ) ( x − x3 ) ( x − x 0 ) ( x − x2 ) ( x − x3 )
f(x) = f ( x0 ) + f ( x1 )
( x0 − x1 ) ( x0 − x2 ) ( x0 − x3 ) ( x1 − x0 ) ( x1 − x2 ) ( x1 − x3 )
( x − x0 ) ( x − x1 ) ( x − x3 ) ( x − x0 ) ( x − x1 ) ( x − x2 )
+ f ( x2 ) + f ( x3 )
( x2 − x0 ) ( x2 − x1 ) ( x2 − x3 ) ( x3 − x0 ) ( x3 − x1 ) ( x3 − x2 )
( x − 1) ( x − 2) ( x − 5) ( x − 0) ( x − 2) ( x − 5) ( x − 0) ( x − 1) ( x − 5)
= (2) + (3) + (12)
(0 − 1) (0 − 2) (0 − 5) (1 − 0) (1 − 2) (1 − 5) (2 − 0) (2 − 1) (2 − 5)
( x − 0) ( x − 1) ( x − 2)
+ (147)
(5 − 0) (5 − 1) (5 − 2)
1 3
m
=− ( x − 1) ( x − 2) ( x − 5) + x ( x − 2) ( x − 5) – 2x(x – 1) (x – 5)
5 4
49
o
+ x ( x − 1) ( x − 2)
20
.c
⇒ f(x) = x3 + x2 – x + 2
Example 4. Find the unique polynomial P(x) of degree 2 such that:
P(1) = 1, P(3) = 27,
Use Lagrange method of interpolation.
n P(4) = 64
ts
ra
Sol. Here, x0 = 1, x1 = 3, x2 = 4
f(x0) = 1, f(x1) = 27, f(x2) = 64
pi
By Lagrange’s interpolation formula

( x − x1 ) ( x − x2 ) ( x − x0 ) ( x − x2 ) ( x − x0 ) ( x − x1 )
P(x) = f ( x0 ) + f ( x1 ) + f ( x2 )
as
( x0 − x1 ) ( x0 − x2 ) ( x1 − x0 ) ( x1 − x2 ) ( x2 − x0 ) ( x2 − x1 )
( x − 3) ( x − 4) ( x − 1) ( x − 4) ( x − 1) ( x − 3)
g
= (1) + (27) + (64)

(1 − 3) (1 − 4) (3 − 1) (3 − 4) (4 − 1) (4 − 3)
.c
1 2 27 2 64 2
= ( x − 7 x + 12) − ( x − 5 x + 4) + ( x − 4 x + 3) = 8x2 – 19x + 12
w
6 2 3
Example 5. The function y = f(x) is given at the points (7, 3), (8, 1), (9, 1) and (10, 9).
w
Find the value of y for x = 9.5 using Lagrange’s interpolation formula.

Sol. We are given
w
x: 7 8 9 10
f(x): 3 1 1 9
Here, x0 = 7, x1 = 8, x2 = 9, x3 = 10
f(x0) = 3, f(x1) = 1, f(x2) = 1, f(x3) = 9
Lagrange’s interpolation formula is
( x − x1 ) ( x − x2 ) ( x − x3 ) ( x − x 0 ) ( x − x2 ) ( x − x3 )
f(x) = f ( x0 ) + f ( x1 )
( x0 − x1 ) ( x0 − x2 ) ( x0 − x3 ) ( x1 − x0 ) ( x1 − x2 ) ( x1 − x3 )
( x − x0 ) ( x − x1 ) ( x − x3 )
+ f ( x2 ) + ( x − x0 ) ( x − x1 ) ( x − x2 ) f ( x3 )
( x2 − x0 ) ( x2 − x1 ) ( x2 − x3 ) ( x3 − x0 ) ( x3 − x1 ) ( x3 − x2 )
www.cgaspirants.com
www.cgaspirants.com
( x − 8) ( x − 9) ( x − 10) ( x − 7) ( x − 9) ( x − 10) ( x − 7) ( x − 8) ( x − 10)

= (3) + (1) + (1)
(7 − 8) (7 − 9) (7 − 10) (8 − 7) (8 − 9) (8 − 10) (9 − 7) (9 − 8) (9 − 10)
( x − 7) ( x − 8) ( x − 9)
+ (9) ...(1)
(10 − 7) (10 − 8) (10 − 9)
Putting x = 9.5 in eqn. (1), we get
1 1
f(9.5) = − (9.5 − 8) (9.5 − 9) (9.5 − 10) + (9.5 − 7) (9.5 − 9) (9.5 − 10)
2 2
1 3
– (9.5 − 7) (9.5 − 8) (9.5 − 10) + (9.5 − 7) (9.5 − 8) (9.5 − 9) = 3.625.
2 2
m
Example 6. Use Lagrange’s interpolation formula to fit a polynomial to the data:
x: –1 0 2 3
o
ux : –8 3 1 12
.c
Hence or otherwise find the value of u1.
Sol. Here,
x0 = – 1,
f(x0) = – 8,
x1 = 0,
f(x1) = 3,
n x2 = 2,
f(x2) = 1,
ts
x3 = 3
f(x3) = 12
ra
Lagrange’s interpolation formula is

( x − x1 ) ( x − x2 ) ( x − x3 ) ( x − x 0 ) ( x − x2 ) ( x − x3 )
f ( x0 ) + f ( x1 )
pi
ux =
( x0 − x1 ) ( x0 − x2 ) ( x0 − x3 ) ( x1 − x0 ) ( x1 − x2 ) ( x1 − x3 )
as
( x − x0 ) ( x − x1 ) ( x − x3 ) ( x − x0 ) ( x − x1 ) ( x − x2 )
+ f ( x2 ) + f ( x3 )
( x2 − x0 ) ( x2 − x1 ) ( x2 − x3 ) ( x3 − x0 ) ( x3 − x1 ) ( x3 − x2 )
g
( x − 0) ( x − 2) ( x − 3) ( x + 1) ( x − 2) ( x − 3) ( x + 1) ( x − 0) ( x − 3)
= (− 8) + (3) + (1)
.c
(− 1 − 0) (− 1 − 2) (− 1 − 3) (0 + 1) (0 − 2) (0 − 3) (2 + 1) (2 − 0) (2 − 3)
( x + 1) ( x − 0) ( x − 2)
w
+ (12)
(3 + 1) (3 − 0) (3 − 2)
w
2 1 1
= x ( x − 2) ( x − 3) + ( x + 1) ( x − 2) ( x − 3) – ( x + 1) x ( x − 3) + ( x + 1) x ( x − 2)
3 2 6
w
⇒ ux = 2x3 – 6x2 + 3x + 3 ...(1)

Putting x = 1 in (1), we get
u1 = 2(1)3 – 6(1)2 + 3(1) + 3 = 2.
Example 7. By means of Lagrange’s formula, prove that
(i) y0 =
1
(y1 + y− 1 ) −
1 1 LM 1
(y3 − y1 ) − (y− 1 − y−3 )
OP
2 8 2 N 2 Q
(ii) y3 = 0.05 (y0 + y6 ) − 0.3 (y1 + y5 ) + 0.75 (y 2 + y4 )
(iii) y1 = y3 – 0.3 (y5 – y–3 ) + 0.2 (y–3 – y–5 ).
www.cgaspirants.com
www.cgaspirants.com
Sol. (i) For the arguments –3, –1, 1, 3, the Lagrange’s formula is
( x + 1) ( x − 1) ( x − 3) ( x + 3) ( x − 1) ( x − 3) ( x + 3) ( x + 1) ( x − 3)
yx = y− 3 + y−1 + y
(−3 + 1) (−3 − 1) (−3 − 3) (−1 + 3) (−1 − 1) (−1 − 3) (1 + 3) (1 + 1) (1 − 3) 1
( x + 3) ( x + 1) ( x − 1)
+ y3 ...(1)
(3 + 3) (3 + 1) (3 − 1)
1 9 9 1
y0 = − y− 3 + y−1 + y1 − y3
16 16 16 16
1
= ( y1 + y− 1 ) −
LM
1 1 1
( y3 − y1 ) − ( y− 1 − y− 3 )
OP
N Q
m
2 8 2 2
(ii) For the arguments 0, 1, 2, 4, 5, 6, the Lagrange’s formula is
o
( x − 1) ( x − 2) ( x − 4) ( x − 5) ( x − 6) ( x − 0) ( x − 2) ( x − 4) ( x − 5) ( x − 6)
yx = y0 + y1
(0 − 1) (0 − 2) (0 − 4) (0 − 5) (0 − 6) (1 − 0) (1 − 2) (1 − 4) (1 − 5) (1 − 6)
.c
( x − 0) ( x − 1) ( x − 4) ( x − 5) ( x − 6) ( x − 0) ( x − 1) ( x − 2) ( x − 5) ( x − 6)
+ y2 + y4
+
(2 − 0) (2 − 1) (2 − 4) (2 − 5) (2 − 6)
( x − 0) ( x − 1) ( x − 2) ( x − 4) ( x − 6)
n
(5 − 0) (5 − 1) (5 − 2) (5 − 4) (5 − 6)
y5 + ts
(4 − 0) (4 − 1) (4 − 2) (4 − 5) (4 − 6)
( x − 0) ( x − 1) ( x − 2) ( x − 4) ( x − 5)
(6 − 0) (6 − 1) (6 − 2) (6 − 4) (6 − 5)
y6 ...(1)
ra
y3 = 0.05 y0 – 0.3 y1 + 0.75 y2 + 0.75 y4 – 0.3 y5 + 0.05 y6
= 0.05 (y0 + y6) – 0.3(y1 + y5) + 0.75 (y2 + y4).
pi
(iii) For the arguments –5, –3, 3, 5, the Lagrange’s formula is

( x + 3) ( x − 3) ( x − 5) ( x + 5) ( x − 3) ( x − 5)
as
yx = y−5 + y−3
(− 5 + 3) (− 5 − 3) (− 5 − 5) (− 3 + 5) (− 3 − 3) (− 3 − 5)
( x + 5) ( x + 3) ( x − 5) ( x + 5) ( x + 3) ( x − 3)
g
+ y3 + y5 ...(1)
(3 + 5) (3 + 3) (3 − 5) (5 + 5) (5 + 3) (5 − 3)
.c
Putting x = 1 in eqn. (1), we get

y1 = – 0.2 y– 5 + 0.5y–3 + y3 – 0.3 y5 = y3 – 0.3 ( y5 – y–3) + 0.2 ( y–3 – y–5).
w
ASSIGNMENT
w
1. Using Lagrange’s interpolation formula, find polynomial which takes the values 3, 12, 15, –21
w
when x has the values 3, 2, 1 and –1. (U.P.T.U. 2014)

2. Values of f(x) for values of x are given as
f(1) = 4, f(2) = 5, f(7) = 5, f(8) = 4
Find f(6) and also the value of x for which f(x) is maximum or minimum.
3. (i) Using Lagrange interpolation formula, calculate f(3) from the following table:
x: 0 1 2 4 5 6
f(x): 1 14 15 5 6 19 (U.P.T.U. 2006)
(ii) Use Lagrange’s interpolation formula to compute f(5.5) from the following data:
x: 0 1 4 5 6
f(x): 1 14 15 6 3 [G.B.T.U. (C.O.) 2011]
www.cgaspirants.com
www.cgaspirants.com
4. Compute f (27) from the following data using Lagrange’s interpolation formula.
x: 14 17 31 35
f(x): 68.7 64.0 44.0 39.1 (U.P.T.U. 2007)
5. (i) Find the value of tan 33° by Lagrange’s formula if
tan 30° = 0.5774, tan 32° = 0.6249, tan 35° = 0.7002, tan 38° = 0.7813.
(ii) Find the value of f(0.55) using Lagrange interpolation with the following table of values:
x: 0.4 0.5 0.7 0.8
f(x): – 0.916 – 0.693 – 0.357 – 0.223 [U.P.T.U. MCA (SUM) 2009]
6. (i) Use Lagrange’s formula to find f(6) from the following table:
x: 2 5 7 10 12
f(x): 18 180 448 1210 2028.
m
(ii) Apply Lagrange’s formula to find f(15), if
x: 10 12 14 16 18 20
o
f(x): 2420 1942 1497 1109 790 540.
.c
7. Derive the Lagrange’s interpolation formula for unequal intervals. Find the value of f(2.6) from
the given table (Use above formula). [U.P.T.U. (MCA) 2006]
x: –2 –1 1 2 6
f(x): –4 14 –4 – 16
8. If y0, y1, ..., y9 are consecutive terms of a series, prove that
1
n ts 196
ra
y5 = [56(y4 + y6) – 28(y3 + y7) + 8(y2 + y8) – (y1 + y9)]
70
9. Using the following table, find f(x) as a polynomial in x:
pi
x: –1 0 3 6 7
f(x): 3 –6 39 822 1611. (U.P.T.U. 2009)
as
10. (i) If y(1) = – 3, y(3) = 9, y(4) = 30, and y(6) = 132, find the four-point Lagrange interpolation
polynomial that takes the same values as the function y at the given points.
(ii) Apply Lagrange’s formula to find a cubic polynomial which approximates the data:
g
x: –2 –1 2 3
.c
y(x): – 12 –8 3 5 [G.B.T.U. (C.O.) 2010]

(iii) Find the unique polynomial P(x) of degree 2 or less such that P(2) = 8, P(4) = 64, P(5) = 125
w
using the Lagrange interpolation formula. Hence evaluate P(2.5).

[G.B.T.U. MCA (SUM) 2010]
w
11. Given the table of values

x: 150 152 154 156
w
y= x: 12.247 12.329 12.410 12.490
Evaluate 155 using Lagrange’s interpolation formula.

12. Use Lagrange’s method to find a polynomial p(x) of lowest possible degree such that p(n) = 2n for
n = 0, 1, 2, 3, 4.
13. Values of f(x) are given at a, b and c. Show that the maximum is obtained by
f (a) (b2 − c2 ) + f (b) (c2 − a2 ) + f (c) (a2 − b2 )
x= .
f (a) (b − c) + f (b) (c − a) + f (c) (a − b)
14. (i) Find a Lagrange’s interpolating polynomial for the data given below:
x0 = 1, x1 = 2.5, x2 = 4 and x3 = 5.5
f(x0) = 4, f(x1) = 7.5, f(x2) = 13 and f(x3) = 17.5
Also, find the value of f(5).
www.cgaspirants.com
www.cgaspirants.com
(ii) Use Lagrange’s method to find interpolating polynomial of degree 3 to fit the following data:
x: 0 1 2 3
f(x): 0 1.7183 6.3891 19.0855
Hence compute the value of f(1.5). [G.B.T.U. (MCA) 2011]
15. Determine by Lagrange’s formula, the percentage number of criminals under 35 years:
Age % number of criminals
under 25 years 52
under 30 years 67.3
under 40 years 84.1
under 50 years 94.4
16. Obtain Lagrange’s interpolatory for the following data:
m
x: 1 3 5 7 10
f(x): 13 31 25 37 101
o
Find the values of f(4) and f(8.5) (U.P.T.U. 2015)
.c
Answers
1. x3
– 9x2
+ 17x + 6 2. 5.66; x = 4.5
3. (i) 10
5. (i) 0.64942084
7. – 17.248
(ii)
(ii)
9.
3.096875
– 0.59721875
n
x4 – 3x3 + 5x2 – 6
ts
4. 49.31046
6. (i) 293.99856 (ii) 1294.8437
ra
1 3 3 2 241
10. (i) x3 – 3x2 + 5x – 6 (ii) y(x) = – x – x + x – 3.9
15 20 60
(iii) P(x) = 11x2 – 38x + 40, 13.75
pi
x4 x 3 11 2 7 x
11. 12.45 − +
12. x + +1
24 12 24 12
as
4 3 14 2 5 115
14. (i) – x + x − x+ ; 16.2962963 (ii) 0.84551x3 – 1.06028x2 + 1.93307x ; 3.367571
27 9 3 27
15. 77.405
g
1
.c
16. f(x) = – (91x4 – 2401x3 + 19571 x2 – 60431 x + 29130) ; f(4) = 27.616, f(8.5) = 64.83.
1080
w
4.36 ERROR IN LAGRANGE’S INTERPOLATION FORMULA

w
Remainder,
Π n + 1 ( x)
w
y(x) – Ln(x) = Rn(x) = y(n + 1) (ξ), a < ξ < b

(n + 1) !
where Lagrange’s formula is for the class of functions having continuous derivatives of order
up to (n + 1) on [a, b].
Quantity EL = max. |Rn(x)| may be taken as an estimate of error.
[a , b]
Let us assume
|y(n + 1) (ξ)| ≤ Mn + 1, a ≤ ξ ≤ b
Mn + 1
then, EL ≤ max. |Πn + 1(x)|.
(n + 1) ! [ a, b]
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Show that the truncation error of quadratic interpolation in an equidistant
h3
table is bounded by max |f ″′(ξ )| where h is the step size and f is the tabulated function.
9 3
Sol. Let xi – 1, xi, xi + 1 denote three consecutive equi-spaced points with step size h.
The truncation error of the quadratic Lagrange interpolation is bounded by
M3
|E2(f ; x)| ≤ max |(x – xi – 1)(x – xi)(x – xi + 1)|
6
m
where xi – 1 ≤ x ≤ xi + 1 and M3 = max |f ″′(x)|
a≤ x≤b
x − xi
o
Substitute t = then,
h
.c
x – xi – 1 = x – (xi – h) = x – xi + h = th + h = (t + 1)h
x – xi + 1 = x – (xi + h) = x – xi – h = th – h = (t – 1)h
and
Setting g′(t) = 0, we get
n ts
(x – xi – 1)(x – xi)(x – xi + 1) = (t + 1) t(t – 1)h3 = t(t2 – 1)h3 = g(t)
ra
1
3t2 – 1 = 0 ⇒ t=± .
3
For both these values of t, we obtain
pi
2h 3
max |(x – xi – 1)(x – xi)(x – xi + 1)| = h3 max |t(t2 – 1)| =
as
3 3
− 1≤ t ≤ 1
Hence, the truncation error of the quadratic interpolation is bounded by

g
h3
M3
.c
|E2(f ; x)| ≤
9 3
h3
w
or, |E2(f ; x)| ≤

max |f ″′(ξ)|.
9 3
w
Example 2. Determine the step size that can be used in the tabulation of f(x) = sin x in
LM π OP at equally spaced nodal points so that the truncation error of the quadratic
w
the interval 0,
N 4Q
interpolation is less than 5 × 10 –8.
Sol. From Ex. 1, we have
h3
|E2(f ; x)| ≤ M3
9 3
For f(x) = sin x, we get f ″′(x) = – cos x and M3 = max |cos x| = 1
0 ≤ x ≤ π/4
Hence the step size h is given by
h3
≤ 5 × 10–8 or h ≈ 0.009
9 3
www.cgaspirants.com
www.cgaspirants.com
FG π IJ from the
Example 3. Using Lagrange’s interpolation formula, find the value of sin
H 6K
following data:
x: 0 π/4 π/2
y = sin x: 0 0.70711 1.0
Also estimate the error in the solution.
FG π − 0IJ FG π − π IJ FG π − 0IJ FG π − π IJ
F π I H 6 K H 6 2 K (0.70711) + H 6 K H 6 4 K (1) = 0.51743
sin G J =
Sol.
H 6 K FG π − 0IJ FG π − π IJ FG π − 0IJ FG π − π IJ
H 4 KH 4 2K H 2 KH 2 4K
m
Now, y(x) = sin x, y′(x) = cos x, y″(x) = – sin x, y′″(x) = – cos x
o
Hence, |y′″ (ξ)| < 1
.c
when x = π/6,
FG π − 0IJ FG π − π IJ FG π − π IJ
|Rn(x)| ≤
H 6 K H 6 4K H 6 2K
n
3! ts
= 0.02392
ra
which agrees with the actual error in problem.
4.37 EXPRESSION OF RATIONAL FUNCTION AS A SUM OF PARTIAL FRACTIONS

pi
3x2 + x + 1
as
Let f(x) =
( x − 1)( x − 2)( x − 3)
2
Consider φ(x) = 3x + x + 1 and tabulate its values for x = 1, 2, 3, we get
g
x: 1 2 3
.c
φ(x) = 3x2 + x + 1: 5 15 31
w
Using Lagrange’s interpolation formula, we get

( x − 2)( x − 3) ( x − 1)( x − 3) ( x − 1)( x − 2)
w
φ(x) = (5) + (15) + (31)

(1 − 2)(1 − 3) −1 2
5 31
w
= (x – 2)(x – 3) – 15 (x – 1)(x – 3) + (x – 1)(x – 2)

2 2
5 15 31
f (x) = – + .
2( x − 1) x − 2 2( x − 3)
4.38 INVERSE INTERPOLATION
The process of estimating the value of x for the value of y not in the table is called inverse
interpolation. When values of x are unevenly spaced, Lagrange’s method is used by
interchanging x and y.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLE
Example. From the given table:
x: 20 25 30 35
y(x): 0.342 0.423 0.5 0.65
Find the value of x for y(x) = 0.390.
Sol. By inverse interpolation formula,
( y − y1 ) ( y − y2 ) ( y − y3 ) ( y − y0 ) ( y − y2 ) ( y − y3 )
x= x0 + x1
( y0 − y1 ) ( y0 − y2 ) ( y0 − y3 ) ( y1 − y0 ) ( y1 − y2 ) ( y1 − y3 )
( y − y0 ) ( y − y1 ) ( y − y3 ) ( y − y0 ) ( y − y1 ) ( y − y2 )
m
+ x2 + x3
( y2 − y0 ) ( y2 − y1 ) ( y2 − y3 ) ( y3 − y0 ) ( y3 − y1 ) ( y3 − y2 )
( 0.39 − 0.423) ( 0.39 − 0.5) ( 0.39 − 0.65)
o
= ( 20)
( 0.342 − 0.423) ( 0.342 − 0.5) ( 0.342 − 0.65)
.c
( 0.39 − 0.342) ( 0.39 − 0.5) ( 0.39 − 0.65)
+ ( 25)
( 0.423 − 0.342) )( 0.423 − 0.5) ( 0.423 − 0.65)
+
n ts
( 0.39 − 0.342) ( 0.39 − 0.423) ( 0.39 − 0.65)
( 0.5 − 0.342) ( 0.5 − 0.423) ( 0.5 − 0.65)
(30)
ra
( 0.39 − 0.342) ( 0.39 − 0.423) ( 0.39 − 0.5)
+ (35)
( 0.65 − 0.342) ( 0.65 − 0.423) ( 0.65 − 0.5)
= 22.84057797.
pi
4.39 DIVIDED DIFFERENCES

as
Lagrange’s interpolation formula has the disadvantage that if another interpolation point
were added, the interpolation co-efficients will have to be recomputed.
g
We therefore seek an interpolation polynomial which has the property that a polynomial
.c
of higher degree may be derived from it by simply adding new terms.

Newton’s general interpolation formula is one such formula and it employs divided
w
differences.
If (x0, y0), (x1, y1), (x2, y2) ...... are given points then the first divided difference for the
w
arguments x0, x1 is defined by

w
y − y0
| y0 = [x0, x1] = 1
Δ
x1 x1 − x0
y2 − y1
Similarly, [x1, x2] = and so on.
x2 − x1
The second divided difference for x0, x1, x2 is defined as
[ x1 , x2 ] − [ x0 , x1 ]
Δ| 2 y = [x , x , x ] =
x1 , x2 0 0 1 2
x2 − x0
Third divided difference for x0, x1, x2, x3 is defined as
[ x1 , x2 , x3 ] − [ x0 , x1 , x2 ]
[x0, x1, x2, x3] = and so on.
x3 − x0
www.cgaspirants.com
www.cgaspirants.com
4.40 PROPERTIES OF DIVIDED DIFFERENCES
1. The divided differences are symmetrical in their arguments i.e., independent of

the order of arguments.
y1 y0
[x0, x1] = + = [x1, x0]
x1 − x0 x0 − x1
y0 y1 y2
Also, [x0, x1, x2] = + +
( x0 − x1 )( x0 − x2 ) ( x1 − x0 )( x1 − x2 ) ( x2 − x0 )( x2 − x1 )
= [x2, x0, x1] or [x1, x2, x0]
2. The nth divided differences of a polynomial of nth degree are constant.
Let the arguments be equally spaced so that
m
x1 – x0 = x2 – x1 = ..... = xn – xn – 1 = h
y1 − y0 Δy0
then, [x0, x1] = =
o
x1 − x0 h
FG Δy IJ = 1 . 1
.c
[ x1 , x2 ] − [ x0 , x1 ] 1 1 Δy0
[x0, x1, x2] = = (Δ2 y0)
( x2 − x0 ) 2h Hh −
h K 2! h 2
In general,
1
.
n ! hn
n 1 n
[x0, x1, x2, ......, xn] = Δ y0
ts
If tabulated function is a nth degree polynomial. ∴ Δny0 = constant
ra
∴ nth divided differences will also be constant.
4.41 NEWTON’S DIVIDED DIFFERENCE INTERPOLATION FORMULA

pi
(U.P.T.U. 2009)
as
Let y0, y1, ......, yn be the values of y = f(x) corresponding to the arguments x0, x1, ......, xn, then
from the definition of divided differences, we have
y − y0
g
[x, x0] =
x − x0
.c
so that, y = y0 + (x – x0) [x, x0] ...(1)

[ x, x0 ] − [ x0 , x1 ]
w
Again, [x, x0, x1] =

x − x1
w
which gives, [x, x0] = [x0, x1] + (x – x1) [x, x0, x1] ...(2)
From (1) and (2), y = y0 + (x – x0) [x0, x1] + (x – x0) (x – x1) [x, x0, x1] ...(3)
w
[ x, x0 , x1 ] − [ x0 , x1 , x2 ]
Also, [x, x0, x1, x2] =
x − x2
which gives [x, x0, x1] = [x0, x1, x2] + (x – x2) [x, x0, x1, x2] ...(4)
From (3) and (4), y = y0 + (x – x0) [x0, x1] + (x – x0) (x – x1) [x0, x1, x2]
+ (x – x0) (x – x1) (x – x2) [x, x0, x1, x2]
Proceeding in this manner, we get
y = f(x) = y0 + (x – x0) [x0, x1] + (x – x0) (x – x1) [x0, x1, x2]
+ (x – x0) (x – x1) (x – x2) [x0, x1, x2, x3] + ..... + (x – x0) (x – x1) (x – x2)
..... (x – xn – 1) [x0, x1, x2, x3, ......, xn] + (x – x0) (x – x1) (x – x2)
..... (x – xn) [x, x0, x1, x2, ......, xn]
which is called Newton’s general interpolation formula with divided differences, the last term
being the remainder term after (n + 1) terms.
www.cgaspirants.com
www.cgaspirants.com
Newton’s divided difference formula can also be written as
y = y0 + (x – x0) Δ| y0 + (x – x0) (x – x1) Δ| 2y0 + (x – x0) (x – x1) (x – x2) Δ| 3y0

+ (x – x0) (x – x1) (x – x2) (x – x3) Δ| 4y0 + ..... + (x – x0) (x – x1) ..... (x – xn–1) Δ| ny0
EXAMPLES
Example 1. (i) Find the third divided difference with arguments 2, 4, 9, 10 of the function
f(x) = x3 – 2x.
1
(ii) If f(x) = 2 , find the first divided differences f(a, b), f(a, b, c), f(a, b, c, d).
x
m
(iii) If f(x) = g(x) h(x), prove that f(x1, x2) = g(x1) h(x1, x2) + g(x1, x2) h (x2).
Sol. (i)
o
x f(x) Δ| f(x) Δ| 2f(x) Δ| 3f(x)
.c
2 4
56 − 4
ts
= 26
4−2
131 − 26
4 56 = 15
n
9−2
711 − 56 23 − 15
ra
= 131 =1
9−4 10 − 2
269 − 131
9 711 = 23
10 − 4
pi
980 − 711
= 269
10 − 9
as
10 980
Hence third divided difference is 1.

g
(ii)
.c
1
x f(x) = Δ| f(x) Δ| 2f(x) Δ| 3f(x)
x2
w
1
a
a2
w
FG 1 − 1 IJ
Hb a K
2 2 FG a + b IJ
w
b−a
= –
Ha b K
2 2
1 ab + bc + ca
b 2
b a2b2 c2
FG b + c IJ FG abc + acd + abd + bcd IJ
−
Hb c K
2 2 –
H 2 2 2 2
a b c d K
1 bc + cd + db
c 2
c b2 c2 d 2
FG c + d IJ
−
Hc d K
2 2
1
d
d2
www.cgaspirants.com
www.cgaspirants.com
From the above divided difference table, we observe that first divided differences,
f(a, b) = –
FG a + bIJ
Ha b K
2 2
f(a, b, c) =
ab + bc + ca
and f(a, b, c, d) = –
FG abc + acd + abd + bcd IJ
a 2 b2 c 2 H 2 2 2
a b c d 2K
h( x2 ) − h( x1 ) g ( x2 ) − g( x1 )
(iii) RHS = g(x1) + h( x2 )
x2 − x1 x2 − x1
1
= [{g(x1) h(x2) – g(x1)h(x1)} + {g(x2) h(x2) – g(x1) h(x2)}]
x2 − x1
m
g ( x2 ) h ( x2 ) − g ( x1 ) h ( x1 )
= = Δ| g(x1) h(x1) = Δ| f(x1) = f(x1, x2) = LHS.
x2 − x1 x2 x2
FG 1 IJ = − 1
o
Δ| 3
Example 2. (i) Prove that bcd H a K abcd
.c
1 LM ( − 1)n OP
(ii) Show that the nth divided differences [x0, x1, ....., xn] for ux =
Sol. (i)
n ts x
is
N
x0 x1 ..... xn
.
Q
ra
x f(x) Δ| f(x) Δ| 2f(x) Δ| 3f(x)
1
a
pi
a
1 1
−
b a =– 1
as
b−a ba
1 1
b (– 1)2
b abc
g
1 1
−
.c
c b =– 1 (– 1)3
1
c−b bc abcd
w
1 1
c (– 1)2
c bdc
w
1 1
−
d c =– 1
d−c dc
w
1
d
d
|3
From the table, we observe that Δ
FG 1IJ = – 1 .
bcd H a K abcd ...(1)
(ii) From (1), we see that
|3
FG 1 IJ = – 1 = (– 1)
Δ
bcd H a K abcd 3 f(a, b, c, d)
∴ In general,
|n
Δ FG 1IJ = (–1) LM (−1) n OP
x0 , x1 , ....., xn
H xK n f (x0, x1, x2, ....., xn) =
Nx 0 x1 x2 ..... xn
.
Q
www.cgaspirants.com
www.cgaspirants.com
Example 3. Using Newton’s divided difference formula, find a polynomial function

satisfying the following data:
x: –4 –1 0 2 5
f(x): 1245 33 5 9 1335
Hence find f(1). (U.P.T.U. 2014)
Sol. The divided difference table is
x f(x) Δ| f(x) Δ| 2f(x) Δ| 3f(x) Δ| 4f(x)

–4 1245
– 404
–1 33 94
m
– 28 – 14
0 5 10 3
2 13
o
2 9 88
.c
442
5 1335
Applying Newton’s divided difference formula

f(x) = 1245 + (x + 4) (– 404) + (x + 4) (x + 1) 94
n ts
+ (x + 4) (x + 1) (x – 0) (– 14) + (x + 4)(x + 1)x(x – 2)(3)
ra
= 3x4 – 5x3 + 6x2 – 14x + 5
Hence, f(1) = 3 – 5 + 6 – 14 + 5 = – 5.
Example 4. Find f ′(10) from the following data:
pi
x: 3 5 11 27 34
as
f(x): – 13 23 899 17315 35606.

Sol. Divided difference table is
g
Δ| f(x) Δ| 2f(x) Δ| 3f(x) Δ| 4f(x)

.c
x f(x)
3 – 13
w
18
5 23 16
w
146 1
11 899 40 0
w
1026 1
27 17315 69
2613
34 35606
By Newton’s divided difference formula,

f(x) = – 13 + (x – 3) 18 + (x – 3)(x – 5)16 + (x – 3)(x – 5)(x – 11)1
∴ f ′(x) = 3x2 – 6x – 7
Put x = 10, f ′(10) = 3(10)2 – 6(10) – 7 = 233.
www.cgaspirants.com
www.cgaspirants.com
Example 5. Given the following table, find f(x) as a polynomial in powers of (x – 5)

x: 0 2 3 4 7 9
f(x): 4 26 58 112 466 922.
Sol. Divided difference table is
x f(x) Δ| f(x) Δ| 2f(x) Δ| 3f(x)

0 4
11
2 26 7
32 1
3 58 11
m
54 1
4 112 16
118 1
o
7 466 22
.c
228
9 922
By Newton’s divided difference formula, we get

f(x) = 4 + (x – 0)(11) + (x – 0)(x – 2)7 + (x – 0)(x – 2)(x – 3)1
= x3 + 2x2 + 3x + 4
n ts
ra
In order to express it in power of (x – 5), we use synthetic division, as
5 1 2 3 4
pi
5 35 190
5 1 7 38 194
as
5 60
5 1 12 98
g
5
.c
1 17
w
∴ 2x2 + x3 + 3x + 4 = (x – 5)3 + 17(x – 5)2 + 98 (x –5) + 194.

Example 6. By means of Newton’s divided difference formula, find the values of f(8)
w
and f(15) from the following table:

w
x: 4 5 7 10 11 13
f(x): 48 100 294 900 1210 2028.
[G.B.T.U. 2011, U.P.T.U. (MCA) 2009]
Sol. Newton’s divided difference formula, using the arguments 4, 5, 7, 10, 11 and 13 is
| f(4) + (x – 4)(x – 5) Δ
f(x) = f(4) + (x – 4) Δ | 2 | 3 f(4) + ...(1)
f(4) + (x – 4)(x – 5)(x – 7) Δ
5 5, 7 5, 7, 10
www.cgaspirants.com
www.cgaspirants.com
The divided difference table is as under:
x f(x) Δ| f(x) Δ| 2 f(x) Δ| 3f(x) Δ| 4f(x)

4 48
100 − 48
= 52
5−4
97 − 52
5 100 = 15
7−4
294 − 100 21 − 15
= 97 = 1
7−5 10 − 4
202 − 97
7 294 = 21 0
10 − 5
m
900 − 294 27 − 21
= 202 =1
10 − 7 11 − 5
o
310 − 202
10 900 = 27 0
11 − 7
.c
1210 − 900 33 − 27
= 310 =1
11 − 10 13 − 7
11 1210
n
2028 − 1210
= 409
ts
409 − 310
13 − 10
= 33
ra
13 − 11
13 2028
Substituting the values of the divided differences in (1),

pi
f(x) = 48 + (x – 4) × 52 + (x – 4)(x – 5) × 15 + (x – 5)(x – 4)(x – 7) × 1

as
Putting x = 8 and 15, we get, f(8) = 448 and f(15) = 3150.

Example 7. Given that log10 2 = 0.3010, log10 3 = 0.4771, log10 7 = 0.8451, find the value
g
of log10 33.
Sol. log 30 = 1.4771, log 32 = 5 log 2 = 5 × 0.3010 = 1.5050
.c
log 36 = 2 (log 2 + log 3) = 2 × (0.3010 + 0.4771) = 1.5562

w
70
log 35 = log = log 70 – log 2 = 1.8451 – 0.3010 = 1.5441.
2
w
Divided difference table is as under:

w
x 104 log10 x 104 Δ| log10 x 104 Δ

| 2 log x
10 104 Δ
| 3log x
10
30 14771
279
= 139.5
2
9.2
32 15050 = − 1.84
5
391 0.48
= 130.3 = − 0.08
3 6
9.3
35 15441 – = − 2.32
7
129
= 121
1
36 15562
www.cgaspirants.com
www.cgaspirants.com
Applying Newton’s divided difference formula, we get

104 log10 x = 14771 + (x – 30) (139.5) + (x – 30)(x – 32) (–1.84)
+ (x – 30)(x – 32)(x – 35)(–0.08)
Putting x = 33
104 log10 33 = 15184.46
∴ log10 33 = 1.5184.
Example 8. The mode of a certain frequency curve y = f(x) is very near to x = 9 and the
values of frequency density f(x) for x = 8.9, 9.0 and 9.3 are respectively equal to 0.30, 0.35 and
0.25. Calculate the approximate value of mode.
Sol. Divided difference table is as under:
x 100 f(x) 100 Δ

| f(x) 100 Δ
| 2f(x)
m
8.9 30
o
5 50
=
0.9 9
.c
350 3500
9.0 35 – = −
9 × 0.4 36
9.3 25
n –
10
0.3
=−
ts
100
3
ra
Applying Newton’s divided difference formula
50 FG
3500 IJ
100 f(x) = 30 + (x – 8.9) × + (x – 8.9)(x – 9) −
H K
pi
9 36
∴ 2
f(x) = – 0.9722x + 17.45833x – 17.597217
as
f ′(x) = – 1.9444 x + 17.45833

17.45833
Putting f ′(x) = 0, we get x= = 8.9788
g
1.9444
Also, f ″(x) = – 1.9444 i.e., (–)ve
.c
∴ f(x) is maximum at x = 8.9788. Hence mode is 8.9788.

w
ASSIGNMENT
w
1. Construct Newton’s interpolation polynomial for the data shown in the following table:
w
x: 0 2 3 4
f(x): 7 11 28 63 [U.P.T.U. MCA (SUM) 2009]
2. (i) Given the values:
x: 5 7 11 13 17
f(x): 150 392 1452 2366 5202
Evaluate f (9) using Newton’s divided difference formula.
(ii) Apply Newton’s divided difference formula to find the value of f(8) if
f(1) = 3, f(3) = 31, f(6) = 223, f(10) = 1011, f(11) = 1343.
3. Obtain the Newton’s divided difference interpolating polynomial and hence find f(6), f(5) and f(8).
x: 3 7 9 10
f (x): 168 120 72 63 [G.B.T.U. 2012, U.P.T.U. 2007]
www.cgaspirants.com
www.cgaspirants.com
4. Given that
x: 1 3 4 6 7
y x: 1 27 81 729 2187
Find y5. Why does it differ from 35 ?
5. Use Newton’s divided difference formula to find the interpolating polynomial and hence evaluate
y(9.5) from the given data:
x: 7 8 9 10
y: 3 1 1 9 [G.B.T.U. 2011 ; G.B.T.U. (C.O.) 2011]
6. (i) The following table is given:
x: 0 1 2 5
y: 2 3 12 147
m
What is the form of the function?
(ii) Develop the divided-difference table from the data given below and obtain the interpolation
polynomial f(x):
o
x: 1 3 5 7 11
.c
f(x): 5 11 17 23 29
Also, find the value of f(19.5). (U.P.T.U. 2009)
7.
8.
x: 5 6
n
9 11
ts
Find the function ux in powers of x – 1 given that u0 = 8, u1 = 11, u4 = 68, u5 = 123.
For the following table, find f(x) as a polynomial in x using Newton’s divided difference formula.
f(x): 12 13 14 16 (U.P.T.U. 2006)

ra
9. (i) Apply Newton’s divided difference method to obtain an interpolatory polynomial for the
following data:
pi
x: 3 5 7 9 11 13
f(x) : 31 51 17 19 90 110 (U.P.T.U. 2015)
as
(ii) Find the Newton’s divided difference interpolation polynomial for:

x: 0.5 1.5 3.0 5.0 6.5 8.0
f(x): 1.625 5.875 31.0 131.0 282.125 521.0
g
10. Using Newton’s divided difference formula, calculate f(6) from the following data:
.c
x: 1 2 7 8
f(x): 1 5 5 4
w
11. (i) Using Newton divided difference method, find the interpolating polynomial and hence compute
f (3) from the following table: [M.T.U. 2014, G.B.T.U. 2013]
w
x: 0 1 2 4 5 6
f(x): 1 14 15 5 6 19
w
(ii) Use Newton’s divided difference method to compute f(5.5) from the following data:
x: 0 1 4 5 6
f(x): 1 14 15 6 3 (G.B.T.U. 2010)
12. Given the data f(1) = 4, f(2) = 5, f(7) = 5, f(8) = 4. Find the the value of f(6) and also the value of x
for which f(x) is maximum or minimum. (M.T.U. 2013)
13. For the following table, find f (x) as a polynomial in x using Newton’s divided difference formula:
x: –1 0 3 6 9
f (x): 3 –6 39 822 1611
Hence compute f(1). (M.T.U. 2012; U.P.T.U. 2009)
14. Using the Newton’s divided difference formula, find a polynomial which takes the values 3, 12,
15, – 21 when x has the values 3, 2, 1, – 1 respectively. (U.P.T.U. 2008)
www.cgaspirants.com
www.cgaspirants.com
15. Find Newton’s divided differences polynomial for the following data:
x: –3 –1 0 3 5
f (x): – 30 – 22 – 12 330 3458 (M.T.U. 2013)
16. Compute f ′(3) from the following table:
x: 1 2 4 8 10
f (x): 0 1 5 21 27 (G.B.T.U. 2011)
17. Find f″′(5) from the data given below:
x: 2 4 9 13 16 21 29
f (x): 57 1345 66340 402052 1118209 4287844 21242820
Answers
m
1. x3 – 2x + 7 2. (i) 810 (ii) 521
3 2
3. x – 21x + 119x – 27; 147, 168, 93 4. 208.82222, 3x is not a polynomial
5. x3 – 23x2 + 174x – 431, 3.625
o
− x4 x3 43 x 2 71 107
.c
6. (i) x3 + x2 – x + 2 (ii) – 112.3955078; + − + x+
320 20 160 20 64
7. (x – 1)3 + 2(x – 1)2 + 4(x – 1) + 11
9.
n ts
8.
x3 7 x2 557 x
20
−
6
+
60
– 11.5
(i) f(x) = – 0.025 x5 + 0.7265625 x4 – 6.3125 x3 + 9.796875 x2 + 79.3375 x – 177.5234375

(ii) x3 + x + 1 10. 6.2381
ra
11. (i) x3 – 9x2 + 21x + 1, 10 (ii) 3.096875 12. 5.66, x = 4.5

257 4 1703 3 377 2 527
pi
13. f(x) = − x + x − x − x − 6 , – 42.037

270 135 30 15
14. f(x) = x3 – 9x2 + 17x + 6 15. 5x4 + 9x3 – 27x2 – 21x – 12
as
16. 1.97916 17. 1626.

g
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
TEST YOUR KNOWLEDGE
1. What is meant by convergence of iterative method? (M.T.U. 2012)

2. Isolate the roots of the equation x3 – 4x + 1 = 0. (M.T.U. 2012)
3. Derive Newton-Raphson formula to find approximate root of the equation f(x) = 0.
(M.T.U. 2013)
4. What is the order of convergence of Newton-Raphson method?
5. Find the missing values in the following table:
x: 0 5 10 15 20 25
y: 6 10 — 17 — 31
m
6. Prove that:
(i) (Δ + 1) (1 – ∇) ≡ 1 (ii) δE1/2 ≡ Δ
o
F 1 + δ2 I 1 / 2 + δ = E Δ ∇
GH 2 4 JK 2
.c
(iii) 1/2 (iv) − = ∇ + Δ.
∇ Δ
7.
8.
9. Verify that ∇E ≡ Δ.
n ts
Construct the forward difference table for f(x) = x3 – 2x2 + 4x + 5 for x = 1, 3, 5, 7. (M.T.U. 2012)
Define central difference operator, shift operator and the average operator. (G.B.T.U. 2012)
(M.T.U. 2012)
FΔ I e
ra
2 x
Ee
10. Prove that: ex = GH E JK x
.
Δ2 e x
(M.T.U. 2013)
pi
11. Show that: hD ≡ sinh–1 (μδ). (M.T.U. 2013)

12. Find the value of Δ2(abcx). (M.T.U. 2013)
as
13. Show that E ≡ 1 + Δ and Δ = ∇ (1 – ∇)–1. [M.T.U. (MCA) 2012]

14. Show that the Regula-Falsi method has linear rate of convergence.
15. Define the order of convergence of an iterative method.
g
16. What do you mean by the rate of convergence of an iterative method? (M.T.U. 2014)
.c
17. Show that δ = E1/2 – E–1/2. (M.T.U. 2014)

18. Define rate of convergence. (M.T.U. 2014)
w
Answers
w
2. 0.2541, 1.860, – 2.1147 4. 2 5. 13.25, 22.5.

12. a(bch – 1)2 bcx
w
www.cgaspirants.com
www.cgaspirants.com
UNIT 5
Numerical Techniques-II
5.1 SOLUTION OF SYSTEM OF SIMULTANEOUS LINEAR EQUATIONS
m
The systems of simultaneous linear equations arise, both directly in modelling physical
o
situations and indirectly in the numerical solution of other mathematical models. Problems
such as determining the potential in certain electrical networks, stresses in a building frame,
.c
flow rates in a hydraulic system etc., are all reduced to solving a set of algebraic equations
simultaneously.
ts
Linear algebraic systems are also involved in the optimization theory, least squares
fitting of data, numerical solution of boundary value problems for ordinary and partial
n
differential equations, statistical inference etc.
ra
Consider a non-homogeneous system of n simultaneous linear algebraic equations in n
unknowns as
a11x1 + a12x2 + a13x3 + ... + a1nxn = b1
pi
a21x1 + a22x2 + a23x3 + ... + a2nxn = b2

a31x1 + a32x2 + a33x3 + ... + a3nxn = b3 ...(1)
as
an1x1 + an2x2 + an3x3 + ... + annxn = bn

g
Using matrix notation, the above system (1) can be written as

.c
AX = B ...(2)
LM
a11 a12 a13 ... a1n OP LM OP
x1 b1LM OP
w
b
a21 a22 a23 ... a2 n
MM PP
where, A = a31 a32 a33 ... a3n , X = x3
x2
MM PP 2
and B = b3
MM PP
MMb PP
w
MMa PP MM x PP
Nn 1 an 2 an3 … ann Q N Q
n nN Q
w
By finding a solution of the system, we mean to obtain the values of n unknowns x1, x2,
x3, ..., xn such that they satisfy the given equations. If B = O, then the system is called
homogeneous.
The methods of solution of linear algebraic eqns. (1) may be classified into two types:
(i) Direct methods. These methods yield the exact solution after a finite number of steps in
absence of round-off errors. In these methods, the amount of computation involved can be
specified in advance. They are independent of the desired accuracy.
(ii) Iterative methods. These methods give a sequence of approximations which
converges when the number of steps tend to infinity.
In some cases, both the direct and iterative methods are combined. First, we may use a
direct method and then the solution may be improved by using iterative methods.
Here, we will study a direct and then an iterative method to solve the system (1).
470
www.cgaspirants.com
www.cgaspirants.com
NUMERICAL TECHNIQUES–II 471
5.2 DIRECT METHODS
Consider a system of linear equations

AX = B ...(1)
where A is coefficient matrix, X is column matrix of unknowns and B is column matrix of
constants. The necessary and sufficient condition for the system (1) to be consistent is
ρ(A) = ρ(A : B)
where ρ(A) denotes the rank of A and ρ(A : B) denotes the rank of augmented matrix (A : B).
Further if ρ(A) = ρ(A : B) = number of unknowns, the system (1) will have unique solution
and if ρ(A) = ρ(A : B) < number of unknowns, the system (1) will have infinite no. of solutions.
Again, the system (1) of equations can also be solved by using the inverse of matrix.
m
If A is non singular square matrix then premultiplying (1) by A–1, we get
A–1 (AX) = A–1B
o
⇒ X = A–1B
.c
adj . A
The inverse of A can be obtained by either A–1 = or by using elementary row
|A|
operations or by using Cayley-Hamilton theorem.
ts
Also, Cramer’s rule can be used to solve (1). But this method is satisfactory only for a
n
small number of simultaneous equations. Although this method involves simple calculation
yet large amount of computation is required to evaluate large order determinants. For an
ra
n × n system, the method involves evaluation of (n + 1) determinants each of nth order. Each
determinant involves large number of multiplications. The addition or subtraction time on
pi
computers is usually small compared to the time taken for multiplications or divisions time.
The method which requires the least multiplication time is the one which is preferred. In
addition to this, there are no simple methods to evaluate large order determinants. For example,
as
to evaluate 10 × 10 determinants, we will have to carry out 10 times evaluation of determinants,

each one of which require the evaluation of 9 × 9 determinants etc. and they naturally need
g
large amount of computation. In view of above, the Cramer’s rule is impractical for large
systems of equations.
.c
Similarly, the computation of inverse of a matrix is the main problem on computers and
it becomes tougher for large system. Hence the method of finding unknowns by X = A–1B is
w
also not practical for large systems.

w
To avoid these unnecessary computations, some simpler and less time consuming
procedures were developed and suggested.
w
5.3 LU DECOMPOSITION METHOD
This method is also known as the Triangularization method or method of factorization.

In this method, the coefficient matrix A of the system of equations AX = B is decomposed
into the product of a lower triangular matrix L and an upper triangular matrix U so that
A = LU ...(1)
LMl 11 0 0 ... 0 OP LMu 11 u12 u13 ... u1n OP

M l
21 l22 0 ... 0
PP MM 0 u22 u23 ... u2 n PP
L = Ml l32 l33 ... 0 U= 0 0 u33 ... u3n
where
MM
31
PP
and
MM PP
Nln1 ln2 ln3 ... lmn Q MN 0 0 0 ... unn PQ
www.cgaspirants.com
www.cgaspirants.com
Using the matrix multiplication and comparing corresponding elements in (1), we obtain
li1 u1j + li2 u2j + ... + lin unj = aij, 1 ≤ j ≤ n
where lij = 0, j > i and uij = 0, i > j
To produce a unique solution, it is convenient to choose either uii = 1 or lii = 1 ; 1 ≤ i ≤ n.
(i) When we choose lii = 1, the method is called the Doolittle’s method.
(ii) When we choose uii = 1, the method is called the Crout’s method.
The given system of equations is
AX = B ...(2)
⇒ LUX = B ...(3)|Using (1)
Let UX = Y then eqn. (3) becomes
m
LY = B ...(4)
The unknowns y1, y2, y3, ..., yn in (4) are determined by forward substitution and the
unknowns x1, x2, x3, ..., xn in UX = Y are obtained by back substitution.
o
Note 1. The method fails if any of the diagonal elements lii or uii is zero.
.c
2. LU decomposition is guaranteed when the matrix A is positive definite.
n EXAMPLES
Example 1. Solve, by Crout’s method, the following system of equations:

ts
ra
x+y+z=3
2x – y + 3z = 16
pi
3x + y – z = – 3. [G.B.T.U. (M. Tech.) 2010]

Sol. We choose uii = 1 and write
as
A = LU
LM1 1 1 OPl11 LM 0 0 OP LM1 u12 u13 OP

g
MN23 −1 3 = l21
PQ MN l22 0
PQ MN00 1 u23
PQ
.c
1 −1 l31 l32 l33 0 1
LMl OP
w
11 l11u12 l11u13
= Ml 21 l21 u12 + l22 l21u13 + l22 u23
PQ
w
Nl
31 l31u12 + l32 l31 u13 + l32 u23 + l33
Equating, we get
w
l11 = 1, l21 = 2, l31 = 3

l11 u12 = 1 ⇒ u12 = 1
l11 u13 = 1 ⇒ u13 = 1
l21 u12 + l22 = – 1 ⇒ l22 = – 3
l31 u12 + l32 = 1 ⇒ l32 = – 2
1
l31 u13 + l22 u23 = 3 ⇒ u23 = –
3
14
l31 u13 + l32 u23 + l33 = – 1 ⇒ l33 = –
3
www.cgaspirants.com
www.cgaspirants.com
Thus, we get
LM OP L 1 1 1OP
A = LU = M
MM
1
2
0
−3
0
0 PP MM0 1 −
1
PP
MN3 −2 −
14
3
PPQ MN0 0
3
1 Q
The given system is
AX = B
⇒ LUX = B ...(1)
Let UX = Y so that (1) becomes
LY = B
m
LM OP
MM21 0 0
PP LM yy OP = LM 163OP
1
o
−3 0
PPQ MN y PQ MN− 3PQ
2
MMN3 14
.c
−2 − 3
3
ts
which gives
y1 = 3
n
2y1 – 3y2 = 16
14
ra
3y1 – 2y2 – y =–3
3 3
10
⇒ y1 = 3, y2 = – , y3 = 4 | By forward substitution
pi
3
Now, UX = Y
1 LM 1 1 x OP LM
3 OP LM OP
as
0 MN00
1 − 1/ 3 y = − 10 / 3
1 z PQ MN
4 PQ MN PQ
g
which gives, x+y+z=3

.c
1 10
y– z=−
3 3
w
z=4
By back substitution
w
x = 1, y = – 2, z = 4.
Example 2. Solve the following system of equations by the LU factorization method:
w
2x + 3y + z = 9
x + 2y + 3z = 6
3x + y + 2z = 8.
Sol. We choose lii = 1 and write
LM2 3 1 OP
1 0 LM 0 OP LMu 11 u12 u13 OP
MN31 2 3 = l21 1
1 2 PQ
l31 l32 MN 0
1 PQ MN 00 u22
0
u23
u33 PQ
LM u 11 u12 u13 OP
= Ml u 21
MNl u
11 l21 u12 + u22 l21 u13 + u23 PP
31 11 l31 u12 + l32 u22 l31 u13 + l32 u23 + u33 Q
www.cgaspirants.com
www.cgaspirants.com
Equating, we get
u11 = 2, u12 = 3, u13 = 1
l21 u11 = 1 ⇒ l21 = 1/2
l31 u11 = 3 ⇒ l31 = 3/2
l21 u12 + u22 = 2 ⇒ u22 = 1/2
l21 u13 + u23 = 3 ⇒ u23 = 5/2
l31 u12 + l32 u22 = 1 ⇒ l32 = – 7
l31 u13 + l32 u23 + u33 = 2 ⇒ u33 = 18
Thus, we get
LM 1 OP LM2 OP
m
0 0 3 1
A = LU =
MN31// 22 1 0
−7 1 PQ MN00 1/ 2 5 / 2
0 18 PQ
o
The given system is
.c
AX = B
⇒
Let
LUX = B
UX = Y so that (1) becomes
LY = B
n ts ...(1)
ra
LM 1 0 0 OP LM y OP LM9OP
1
MN31// 22 1 0
−7 1 PQ MN yy PQ = MN68PQ
2
pi
3
which gives
as
y1 = 9
1
y + y2 = 6
g
2 1
.c
3
y – 7 y2 + y3 = 8
2 1
w
⇒ y1 = 9, y2 = 3/2, y3 = 5 | By forward substitution

Now, UX = Y
w
LM2 3 1 OP LM xOP LM 9 OP
MN00 1/ 2 5 / 2
PQ MN zyPQ = MN35/ 2PQ
w
0 18
which gives,
2x + 3y + z = 9
1 5 3
y+ z=
2 2 2
18 z = 5
By back substitution
35 29 5
⇒ x= , y= , z=
18 18 18
Note. We may also choose uii = 1 to get the solution.
www.cgaspirants.com
www.cgaspirants.com
Example 3. Show that the LU decomposition method fails to solve the system of equations
x1 + x2 – x3 = 2
2x1 + 2x2 + 5x3 = –3
3x1 + 2x2 – 3x3 = 6
whose exact solution is (1, 0, – 1).
Sol. Case I. If we write (taking lii = 1)
LM1 1 −1 OP
1 0 LM0 OP LMu 11 u12 u13 OP
MN23 2 5 = l21 1
2 −3 PQ 0
MN
l31 l32 1 PQ MN00 u22 u23
0 u33 PQ
We obtain,
m
u11 = 1, u12 = 1, u13 = –1
l21 u11 = 2 ⇒ l21 = 2
o
l21 u12 + u22 = 2 ⇒ u22 = 0
.c
Hence LU decomposition method fails as the pivot u22 = 0.
Case II. If we write (taking uii = 1)
LM1
MM2
1 −1
2
OP LMl
5P = Ml
n 11
21
0
l22
0
0
OP LM1
PP MM0
u12
1
tsu13
u23
OP
PP
N3 2 − 3PQ MNl l32 l33 Q N0 0 1 Q
ra
31
We obtain,
l11 = 1, l21 = 2, l31 = 3
pi
l11 u12 = 1 ⇒ u12 = 1

as
l11 u13 = –1 ⇒ u13 = –1

l21 u12 + l22 = 2 ⇒ l22 = 0
Hence LU decomposition method fails as again the pivot l22 = 0.
g
.c
ASSIGNMENT
w
1. Solve the following system(s) of linear equations by Crout’s method.

w
(i) x1 + x2 + x3 = 1 (ii) 10x + y + z = 12

4x1 + 3x2 – x3 = 6 2x + 10y + z = 13
w
3x1 + 5x2 + 3x3 = 4 2x + 2y + 10z = 14

(iii) 3x – y + 2z = 12 (iv) x 1 + x 2 + x3 = 1
x + 2y + 3z = 11 3x1 + x2 – 3x3 = 5
2x – 2y – z = 2 x1 – 2x2 – 5x3 = 10 (M.T.U. 2013)
[M.T.U. 2012, G.B.T.U. 2011, 2013]
LM 2 1 −4 1 OP LM x OP LM 4OP
1
−4 3 5 −2
(v) M
MN 11 −1 1 −1 PP MM xx PP = MM− 102PP .
2
3
(U.P.T.U. 2007)
3 −3 2 Q N x Q N − 1Q
4
www.cgaspirants.com
www.cgaspirants.com
2. Solve the following system of linear equations by LU decomposition method.

(i) x1 + 2x2 + 3x3 = 14 LM
(ii) 2 1 1 −2 OP LM x OP LM−10OP
1
2x1 + 5x2 + 2x3 = 18
3x1 + x2 + 5x3 = 20
2
3
MM43 0
2
2 1
2 0 PP MM xx PP = MM 87PP (U.P.T.U. 2009)
(iii) 5x – 2y + z = 4 4 N1 3 2 −1 Q N x Q N − 5Q
7x + y – 5z = 8
3x + 7y + 4z = 10. (G.B.T.U. 2012)
3. Solve the following system of equations by triangularization method.
(i) 2x – 3y + 10z = 3 (ii) x1 + 2x2 – x3 = 3
– x + 4y + 2z = 20 x1 – x2 + x3 = – 1
5x + 2y + z = – 12 2x1 – 2x2 + 3x3 = 2
m
(iii) x1 + x 2 – x 3 = 2
2x1 + 3x2 + 5x3 = – 3
o
3x1 + 2x2 – 3x3 = 6.
.c
4. Obtain the LU decomposition of the matrix
LM 2 −6 10 OP
so that for 1 ≤ i ≤ 3,
MM 1
N− 1
5
15
n 1
− 5PQ
P ts
ra
(i) uii = 2 (ii) lii = 2.
LM4 1 1 OP
pi
5. Decompose A =
MN31 4 − 2 in the form LU where L is lower triangular matrix and U is the
2 −4 PQ
as
upper triangular matrix and hence solve the system of equations:

4x1 + x2 + x3 = 4
g
x1 + 4x2 – 2x3 = 4
3x1 + 2x2 – 4x3 = 6 (U.P.T.U. 2014)
.c
Answers
w
1. (i) x1 = 1, x2 = 0.5, x3 = – 0.5 (ii) x = 1, y = 1, z = 1

w
(iii) x = 3, y = 1, z = 2 (iv) x1 = 6, x2 = – 7, x3 = 2
(v) x1 = 1, x2 = – 1, x3 = – 1, x4 = – 1
w
2. (i) x1 = 1, x2 = 2, x3 = 3 (ii) x1 = 5, x2 = 6, x3 = – 10, x4 = 8

(iii) x = 1.11926, y = 0.8685, z = 0.14067
3. (i) x = – 4, y = 3, z = 2 (ii) x1 = – 1, x2 = 4, x3 = 4
(iii) x1 = 1, x2 = 0, x3 = – 1
L 1 0 0 OP
2 − 6 10 LM OP LM 2 0 0 OP LM1 −3 5 OP
4. (i) L = M 1/ 2 4 0 ,U= 0 2 −1 MM PP (ii) L = M 1 2 0 P , U = M0 4 −2 P
MN− 1/ 2 6 3 0
PQ
0 2 N Q MN− 1 3 2PQ MN0 0 3PQ
5. x1 = 1, x2 = 0.5, x3 = – 0.5
www.cgaspirants.com
www.cgaspirants.com
5.4 CHOLESKY METHOD
This method is also called square-root method. Consider a system of equations

AX = B ...(1)
If the coefficient matrix A is symmetric and positive definite then A can be decomposed
as
A = LLT ...(2)
where L is a lower triangular matrix.
A may also be decomposed as A = UUT where U is an upper triangular matrix.
From (1) and (2),
m
LLTX = B
⇒ LY = B ...(3)
o
where L TX =Y ...(4)
.c
The values yi, 1 ≤ i ≤ n can be obtained by forward substitution and the solution xi,
1 ≤ i ≤ n are obtained by back substitution.
Example. Solve the following system of equations
LM 1 2 3 OP LM x OP LM 5OP
1
n ts
MN32 8 22
PQ MN xx PQ = MN−106 PQ
ra
2
22 82 3
using Cholesky method.

pi
Sol. Let A = LLT |∵ A is symmetric
LM 1 OP LM OP LMl OP
as
2 3 l11 0 0 11 l21 l31

MN23 8 22 = l21
22 82 PQ
l31 MN l22
l32
0
l33 PQ MN 00 l22
0
l32
l33 PQ
g
LMl OP
.c
2
11 l11l21 l11l31
= Ml 21l11 l212 + l22 2 l21l31 + l22 l32 PP
w
MNl
31l11 l31l21 + l32 l22 l312 + l32 2 + l33 2 Q
w
Equating the corresponding elements, we get

l112 = 1 ⇒ l11 = 1
w
l11 l21 = 2 ⇒ l21 = 2

l11 l31 = 3 ⇒ l31 = 3
2 2
l21 + l22 = 8 ⇒ l22 = 2
l31 l21 + l32 l22 = 22 ⇒ l32 = 8
2 2 2
l31 + l32 + l33 = 82 ⇒ l33 = 3
LM1 0 0 OP
Hence, L = M2 2 0 P
MN3 8 3PQ
www.cgaspirants.com
www.cgaspirants.com
The given system can be written as LLT X = B

or LY = B where LTX = Y
From LY = B, we get
LM
1 0 0 y1 OP LM
5 OP LM OP
2 2 0 y2 = 6
MN
3 8 3 y3 − 10PQ MN PQ MN PQ
⇒ y1 = 5, y2 = – 2, y3 = – 3 |By forward substitution
From LTX = Y, we get
LM
1 2 3 x1 OP LM
5 OP LM OP
0 2 8 x2 = − 2
MN
0 0 3 x3 PQ MN
−3 PQ MN PQ
m
⇒ x1 = 2, x2 = 3, x3 = – 1. | By backward substitution
o
ASSIGNMENT
.c
1. Solve the system of equations
4x1 – x2 = 1
– x1 + 4x2 – x3 = 0
– x2 + 4x3 = 0
n ts
by the Cholesky method.
ra
Answer
pi
15 1 1
1. x1 = , x2 = , x3 = .
56 14 56
as
5.5 ITERATIVE METHODS

g
The iterative or indirect methods start from an approximation to the true solution and if
convergent, derive a sequence of closer approximations. The cycle of computations is repeated
.c
till the desired accuracy is attained. In these methods, the amount of computation depends on
the accuracy required.
w
Generally, direct methods are preferred for the solution of a linear system of equations
but in case of matrices having large number of zero elements, iterative methods are preferred
w
since they preserve those elements.

The direct method discussed so far involve many subtractions. When the terms involved
w
in subtractions are nearly equal, their difference is nearly zero and hence causes inaccuracies.
The inaccuracies due to this inherent weakness of the direct methods cannot be completely
avoided whereas the iterative methods are free from such inaccuracies.
The disadvantage of the iterative methods is that they can not be applied to all systems
of equations. The sufficient condition for their use is that the system of equations should be
diagonally dominant. In other words, after rearranging the equations, if necessary, the larger
coefficients must be along the leading diagonal of the coefficient matrix. Successful use of
iteration process requires that the moduli of diagonal coefficients of given systems should be
large in comparison with the moduli of nondiagonal coefficients.
Among the iterative methods, we will first discuss Jacobi’s method and then Gauss-
Seidel method. It is known that for a given system of equations, the Gauss-Seidel method
converges more rapidly than the Jacobi’s method.
www.cgaspirants.com
www.cgaspirants.com
5.6 JACOBI’S ITERATIVE METHOD OR METHOD OF SIMULTANEOUS

DISPLACEMENTS
Consider a system of n simultaneous linear equations in n unknowns as

a11x1 + a12x2 + a13x3 + ... + a1nxn = b1 U|
a21x1 + a22x2 + a23x3 + ... + a2nxn = b2 |V ...(1)
||
an1x1 + an2x2 + an3x3 + ... + ann xn = bn W
where aii is the largest coefficient in the ith equation (1 ≤ i ≤ n).
System (1) of equations can be rewritten as
U|
m
b1 a a a
x1 = – 12 x2 – 13 x3 – ... – 1n xn
a11 a11 a11 a11
||
o
b2 a a a
– 21 x1 – 23 x3 – ... – 2n xn
V|
x2 = ...(2)
.c
a22 a22 a22 a22
xn =
bn
ann
a a
– n1 x1 – n2 x2 – ... –
ann ann
n an(n − 1)
ann
xn–1
ts ||
W
ra
Let the first approximations to the unknowns x1, x2, ..., xn be x1(1), x2(1), ..., xn(1). Put the
first approximation in RHS of (2) to get the results as x1(2), x2(2), x3(2), ..., xn(2), a system of
pi
second approximations.
This entire process is repeated until the values of x1, x2, ..., xn are obtained to desired
accuracy i.e., until successive values of each of the unknowns are sufficiently close.
as
Jacobi’s method, if suitable to a system of equations, converges for any value of initial
approximations. Generally they are taken as x1(0) = 0, x2(0) = 0, x3(0) = 0, ... and so on in absence
g
of any better estimates.

.c
Example. Solve the following system of equations by Jacobi’s method:

8x – 3y + 2z = 20
w
6x + 3y + 12z = 35
4x + 11y – z = 33.
w
Sol. The coefficient matrix of the given system is not diagonally dominant. Hence we
rearrange the equations such that the elements in the coefficient matrix are diagonally
w
dominant.
8x – 3y + 2z = 20 U|
4x + 11y – z = 33 V| ...(1)
6x + 3y + 12z = 35 W
Now, we write the equations in the form
1
x = (20 + 3y – 2z)
8
U|
y=
1
(33 – 4x + z)
|V ...(2)
11
1
||
z=
12
(35 – 6x – 3y) W
www.cgaspirants.com
www.cgaspirants.com
We start from an initial approximation x(0) = 0, y(0) = 0 and z(0) = 0. Substituting them in
(2), we get
First approximation
1
x(1) = [20 + 3(0) – 2(0)] = 2.5
8
1
y(1) = [33 – 4(0) + (0)] = 3
11
1
z(1) =
[35 – 6(0) – 3(0)] = 2.9166667
12
Second approximation
m
1
x(2) = [20 + 3y(1) – 2z(1)] = 2.895833
8
o
1
y(2) = [33 – 4x(1) + z(1)] = 2.3560606
.c
11
1
Third approximation
z(2) =
12
n ts
[35 – 6x(1) – 3y(1)] = 0.9166666
1
ra
x(3) = [20 + 3y(2) – 2z(2)] = 3.1543561

8
1
pi
y(3) = [33 – 4x(2) + z(2)] = 2.030303

11
1
as
z(3) = [35 – 6x(2) – 3y(2)] = 0.8797348

12
Proceeding in this way, after 12th approximation, we find that the two successive
g
approximations are same upto four decimal places. Hence at this stage, we get
.c
x = 3.0167, y = 1.9858 and z = 0.9118.

w
ASSIGNMENT
w
1. Solve the following system of equations by Jacobi iteration method:

w
(i) 3x + 4y + 15z = 54.8 (ii) 54x + y + z = 110

x + 12y + 3z = 39.66 2x + 15y + 6z = 72
10x + y – 2z = 7.74 – x + 6y + 27z = 85
(iii) 10x1 + 2x2 + x3 = 9 (iv) 30x – 2y + 3z = 75
x1 + 10x2 – x3 = – 22 x + 17y – 2z = 48
– 2x1 + 3x2 + 10x3 = 22 x + y + 9z = 15
(v) 10x1 – x2 = 17
– x1 + 10x2 – x3 = 24
– x1 + 10x3 – x4 = 32
– x3 + 10x4 = 46
www.cgaspirants.com
www.cgaspirants.com
2. Solve the following system of equations using Jacobi’s method:

5x – y + z = 10
2x + 4y = 12
x + y + 5z = – 1
Start with the solution (2, 3, 0).
Answers
1. (i) x = 1.074, y = 2.524, z = 2.765 (ii) x = 1.926, y = 3.573, z = 2.425
(iii) x1 = 1, x2 = – 2, x3 = 3 (iv) x = 2.580, y = 2.798, z = 1.069
(v) x1 = 2, x2 = 3, x3 = 4, x4 = 5 2. x = 2.555, y = 1.722, z = – 1.05
m
5.7 GAUSS-SEIDEL ITERATIVE METHOD OR METHOD OF SUCCESSIVE
DISPLACEMENTS
o
Consider a system of n simultaneous linear eqns. in n unknowns as
U|
.c
a11x1 + a12x2 + a13x3 + ... + a1nxn = b1
a21x1 + a22x2 + a23x3 + ... + a2nxn = b2 |V ...(1)
n
an1x1 + an2x2 + an3x3 + ...... + annxn = bn
||
W
ts
where aii is the largest coefficient in the ith equation (1 ≤ i ≤ n).
ra
We solve each equation of the system (1) for the unknown with the largest coefficient
interms of the remaining unknown. System (1) can be written as
pi
x1 =
b1 a a a
– 12 x2 – 13 x3 – ... – 1n xn
U|
a11 a11 a11 a11
||
as
x2 =
b2 a a a
– 21 x1 – 23 x3 – ... – 2n xn |
a22 a22 a22 a22 V|
g
|| ...(2)
.c
bn a a an(n −1) ||
w
– n1 x1 – n2 x2 – ... –
xn =
ann ann ann ann
xn–1
W
w
In the first equation of (2), we substitute the first approximation x1(1), x2(1), x3(1), ..., xn(1)
into RHS. and denote the result as x1(2).
w
In the second equation, we substitute x1(2), x2(1), x3(1), ..., xn(1) and denote the result as x2(2).
In the third equation, we substitute x1(2), x2(2), x3(1), ..., xn(1) and denote the result as x3(2).
In this manner, we complete the first stage of iteration and the entire process is recpeated
till the values of x1, x2, x3, ..., xn are obtained to the accuracy required.
This method uses an improved component as soon as it is available therefore it is also
called the method of successive displacements.
For any choice of first approximation xj(1) (1 ≤ j ≤ n), Gauss-Seidel method converges if
n aij
every equation of (2) satisfies the condition ∑ aii
≤ 1, (1 ≤ i ≤ n), where ‘<’ sign should be
j =1, j ≠ i
valid in case of at least one equation.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Use Gauss-Seidel iterative method to solve the following system of simulta-
neous equations:
9x + 4y + z = –17
x – 2y – 6z = 14
x + 6y = 4
Perform four iterations. (U.P.T.U. 2014)
Sol. The given system is not diagonally dominant. Hence rearranging the equations as
9x + 4y + z = –17
x + 6y = 4
m
x – 2y – 6z = 14
Now, Gauss-Seidel’s iterative method can be applied.
o
From the above equations, we get
.c
1
x = (– 17 – 4y – z) ...(1)
9
1
y = (4 – x)
6
1
n
z = (x – 2y – 14)
ts ...(2)
...(3)
6
ra
First approximation
Starting with y = 0 = z, we obtain
pi
17
x(1) = – = – 1.8888
9
as
Now, putting x = – 1.888 and z = 0 in eqn. (2), we get

y(1) = 0.9815
g
Again putting x = – 1.888 and y = 0.9815 in eqn. (3), we get

.c
z(1) = – 2.9753
w
1
x(2) = {–17 – 4y(1) – z(1)} = –1.9945
9
w
1
y(2) = {4 – x(2)} = 0.9991
w
6
1 (2)
z(2) = {x – 2y(2) – 14} = – 2.9988
6
Third approximation
x(3) = –1.9997
y(3) = 0.9999
z(3) = – 2.9999
Fourth approximation
x(4) = – 1.9999 ≈ –2
y(4) = 0.9999 ≈ 1
z(4) = – 2.9999 ≈ –3
www.cgaspirants.com
www.cgaspirants.com
Hence after four iterations, we obtain

x = – 2, y = 1, z = – 3.
Example 2. Solve the following system of equations using Gauss-Seidel iterative method:
2x + 10y + z = 51
10x + y + 2z = 44
x + 2y + 10z = 61 [U.P.T.U. (MCA) 2009]
Sol. The given system is not diagonally dominant. Hence rearranging the equations as
10x + y + 2z = 44
2x + 10y + z = 51
x + 2y + 10z = 61
m
Now, Gauss-Seidel’s iterative method can be applied.
From the above equations, we get
o
1
x= (44 − y − 2 z) ...(1)
10
.c
1
y= (51 – 2x – z) ...(2)
z=
10
1
10
n
(61 – x – 2y)
ts ...(3)
ra
First approximation.
Starting with y = 0 = z, we obtain from (1),
pi
x(1) = 4.4
Now, putting x = 4.4 and z = 0 in equation (2), we get
as
y(1) = 4.22
Again, putting x = 4.4 and y = 4.22 in equation (3), we get
z(1) = 4.816
g
Second approximation.
.c
x(2) = 3.0148
y(2) = 4.01544
w
z(2) = 4.995432
w
Third approximation
x(3) = 2.9993696
w
y(3) = 4.00058288
z(3) = 4.999946464
Fourth approximation.
x(4) = 2.999952419 ≈ 3
y(4) = 4.00001487 ≈ 4
z(4) = 5.000001784 ≈ 5
Hence after four iterations, we obtain
x = 3, y = 4 and z = 5
www.cgaspirants.com
www.cgaspirants.com
Example 3. Solve the following system of equations by Gauss-Seidel iterative method:

20x + y – 2z = 17
3x + 20y – z = – 18
2x – 3y + 20z = 25 [M.T.U. 2013; U.P.T.U. 2009]
Sol. The given system is diagonally dominant hence, from the given equations, we obtain
1
x= (17 − y + 2 z) ...(1)
20
1
y= (–18 – 3x + z) ...(2)
20
m
1
z= (25 – 2x + 3y) ...(3)
20
o
First Approximation
Starting with y = 0 = z, we obtain from (1),
.c
x(1) = 0.85
y(1) = –1.0275
n ts
Now, putting x = 0.85, z = 0 in equation (2), we get
Again, putting x = 0.85 and y = – 1.0275 in eqn. (3), we get

ra
z(1) = 1.010875
Second Approximation
pi
x(2) = 1.0024625
y(2) = – 0.999825625
as
z(2) = 0.999779906
Third Approximation
g
x(3) = 0.999969271 ≈ 1
y(3) = –1.000006395 ≈ –1
.c
z(3) = 1.000002114 ≈ 1
w
Hence after three approximations, we obtain,

x = 1, y = – 1, z = 1
w
ASSIGNMENT
w
1. Test if the following system of equations is diagonally dominant and hence solve this system
using Gauss-Seidel method:
2x1 + x2 + 4x3 = 7
3x1 + x2 + 2x3 = 6
– x1 + 4x2 + 2x3 = 5 (G.B.T.U. 2011)
2. Define diagonally dominant system of equations. Solve the following system of equations using
Gauss-Seidel method:
10x + 15y + 3z = 14
– 30x + y + 5z = 17
x + y + 4z = 3 (G.B.T.U. 2012)
www.cgaspirants.com
www.cgaspirants.com
3. (i) Obtain the solution of the following system of equations using Gauss-Seidel iterative method:
2x – 7y – 10z = – 17
5x + y + 3z = 14
x + 10y + 9z = 7
Perform three iterations. [G.B.T.U. (MCA) 2011]
(ii) Solve the following system of linear equations using Gauss-Seidel method:
10x +3y + 7z = 41
3x + 20y + 17z = 101
x + 19y + 23z = 201.
Perform three iterations. (U.P.T.U. 2015)
4. Solve by Gauss-Seidel iterative method:
m
(i) 10x + 2y + z = 9 (ii) 3x1 + 2x2 – x3 = 7
2x + 20y – 2z = – 44 x1 – 3x2 + 2x3 = 4
o
– 2x + 3y + 10z = 22 – x1 + x2 – 3x3 = – 1
(iii) x + 0.01y – 0.02z = 3.3354 (iv) 10x1 – x2 + 2x3 = 4
.c
0.02x + y – 0.05z = 4.8241 x1 + 10x2 + x3 = 7
ts
0.03x – 0.01y + z = 7.341 2x1 + 3x2 + 20x3 = 13
(v) 6x1 – 3x2 + x3 = 11 (vi) 8x – 3y + 2z = 20
n
x1 – 7x2 + x3 = 10 6x + 3y + 12z = 35
2x1 + x2 – 8x3 = – 15 4x + 11y – z = 33
ra
(G.B.T.U. 2013, 2011) [G.B.T.U. (M. Tech.) 2010]

(vii) 27x + 6y – z = 85 (viii) 83x + 11y – 4z = 95
pi
6x + 15y + 2z = 72 7x + 52y + 13z = 104

x + y + 54z = 110 (U.P.T.U. 2015) 3x + 8y + 29z = 71 (M.T.U. 2014)
as
5. Describe a method for solving a system of linear equations. Solve the following system of linear
equations using Gauss-Seidel method:
23x1 + 13x2 + 3x3 = 29
g
5x1 + 23x2 + 7x3 = 37

.c
11x1 + x2 + 23x3 = 43. (U.P.T.U. 2009)

6. Solve the following equations by Gauss-Seidel procedure:
w
9x1 + 2x2 + 4x3 = 20

x1 + 10x2 + 4x3 = 6
w
2x1 – 4x2 + 10x3 = – 15

The answer should be correct to 3 significant digits. [U.P.T.U. MCA (SUM) 2009]
w
7. Using Gauss-Seidel method, solve the following system of linear algebraic equations
10x1 – 2x2 – x3 – x4 = 3
– 2x1 + 10x2 – x3 – x4 = 15
– x1 – x2 + 10x3 – 2x4 = 27
– x1 – x2 – 2x3 + 10x4 = – 9
correct up to four decimal places. Perform seven iterations.
Answers
1. x1 = 1, x2 = 1, x3 = 1 2. x = – 0.4328, y = 1.1055, z = 0.5818
3. (i) x = 1, y = – 3, z = 4 (ii) x = – 3.9429, y = – 6.4887, z = 14.2708
4. (i) x = 1, y = – 2, z = 3 (ii) x1 = 2.72, x2 = – 1.04, x3 = – 0.92
www.cgaspirants.com
www.cgaspirants.com
(iii) x = 3.429, y = 5.119, z = 7.289 (iv) x1 = 0.3567, x2 = 0.6120, x3 = 0.5225

(v) x1 = 1, x2 = – 1, x3 = 2 (vi) x = 3.016, y = 1.985, z = 0.9118
(vii) x = 2.425, y = 3.573, z = 1.925 (viii) x = 1.057, y = 1.367, z = 1.961.
5. x1 = 0.4765, x2 = 1.0189, x3 = 1.5973 6. x1 = 2.74, x2 = 0.987, x3 = – 1.65
7. x1 = 1, x2 = 2, x3 = 3, x4 = 0
5.8 INTRODUCTION TO NUMERICAL DIFFERENTIATION AND INTEGRATION
Consider a function of a single variable y = f(x). If f(x) is defined as an expression, its deriva-
tive or integral may often be determined using the techniques of calculus.
However, when f(x) is a complicated function or when it is given in a tabular form, we
use numerical methods.
m
Here, we will discuss numerical methods for approximating the derivative(s) f (r)(x),
z b
o
r ≥ 1 of a given function f(x) and for the evaluation of the integral f ( x) dx where a, b may be
a
.c
finite or infinite.
The accuracy attainable by these methods would depend on the given function and the
ts
order of the polynomial used. If the polynomial fitted is exact then the error would be
theoretically zero. In practice, however, rounding errors will introduce errors in the calculated
values.
n
ra
The error introduced in obtaining derivatives is in general much worse than that intro-
duced in determining integrals. It may be observed that any errors in approximating a function
are amplified while taking the derivative whereas they are smoothed out in integration.
pi
Thus numerical differentiation should be avoided if an alternative exists.

5.9 NUMERICAL DIFFERENTIATION [M.T.U. 2012, U.P.T.U. 2008]
as
In the case of numerical data, the functional form of f(x) is not known in general. First we
have to find an appropriate form of f(x) and then obtain its derivatives. So “Numerical
g
differentiation” is concerned with the method of finding the successive derivatives of a function
.c
at a given argument, using the given table of entries corresponding to a set of arguments,
equally or unequally spaced. Using the theory of interpolation, a suitable interpolating
w
polynomial can be chosen to represent the function to a good degree of approximation in the
given interval of the argument.
w
For the proper choice of interpolation formula, the criterion is same as in case of
interpolation problems. In case of equidistant values of x, if the derivative is to be found at a
w
point near the beginning or the end of the given set of values, we should use Newton’s forward
or backward difference formula accordingly. Also if the derivative is to be found at a point
near the middle of the given set of values, then we should use any one of the central difference
formulae. However, if the values of the function are not known at equidistant values of x, we
shall use Newton’s divided difference or Lagrange’s formula.
5.10 FORMULAE FOR DERIVATIVES
(1) Newton’s forward difference interpolation formula is

u(u − 1) 2 u(u − 1)(u − 2) 3
y = y0 + u Δy0 + Δ y0 + Δ y0 + ... ...(1)
2! 3!
x−a
where u= ...(2)
h
www.cgaspirants.com
www.cgaspirants.com
Differentiating eqn. (1) with respect to u, we get

dy 2u − 1 2 3u 2 − 6u + 2 3
= Δy0 + Δ y0 + Δ y0 + ... ...(3)
du 2 6
Differentiating eqn. (2) with respect to x, we get
du 1
= ...(4)
dx h
We know that
dy dy du 1 2u − 1 2 LM
3u 2 − 6u + 2 3 FG IJ F I OP
= . =
dx du dx h
Δy0 +
2
Δ y0 +
6 MN H
Δ y0 + ...
K GH JK PQ ...(5)
dy
m
Expression (5) provides the value of at any x which is not tabulated.
dx
Formula (5) becomes simple for tabulated values of x, in particular when x = a and u = 0
o
Putting u = 0 in (5), we get
.c
FG dy IJ 1 1LM 1 1 1
Δy0 − Δ2 y0 + Δ3 y0 − Δ4 y0 + Δ5 y0 − ...
OP ...(6)
H dx K =
h 2 N 3 4 5 Q
x=a

n ts (M.T.U. 2013)
FG IJ = d FG dy IJ du
ra
d2 y d dy
dx 2
=
dx dx H K du H dx K dx
1L F 6u − 18u + 11I Δ y + ...OP 1
pi
2
= M Δ y + (u − 1) Δ y + G
H 12 JK
2 3 4
h MN PQ h
0 0 0
1 L F 6u − 18u + 11I Δ y + ...OP

as
2
=
h MN2 M Δ y + (u − 1) Δ y + G
2
H 12 JK
0
3
PQ
0
4
0 ...(7)
g
Putting u = 0 in eqn. (7), we get

.c
F d yI
2
1 FG Δ y 11 4 IJ
GH dx JK =
H
2
− Δ3 y0 + Δ y0 − ...
K ...(8)
w
2 0
x=a
h2 12
w
Similarly, we get
F d yI FG Δ y IJ
w
3
1 3 4
GH dx JK
3
x=a
=
h 3 H
3
0 −
2
Δ y0 + ...
K ...(9)
and so on.
Formulae for computing higher derivatives may be obtained by successive differentiation.
Aliter: We know that
E = ehD ⇒ 1 + Δ = e hD
Δ2 Δ3 Δ4
∴ hD = log (1 + Δ) = Δ −
+ − + ...
2 3 4
⇒ D=
1 1 1LM 1
Δ − Δ2 + Δ3 − Δ4 + ...
OP [U.P.T.U. (MCA) 2007]
h 2 3 N 4 Q
www.cgaspirants.com
www.cgaspirants.com
D2 =
1FG Δ − 1 Δ + 1 Δ − 1 Δ 2 3 4 IJ 2
1 FG Δ 2
− Δ3 +
11 4 5 5 IJ
h H K H K
Similarly, 2
+ ... = 2
Δ − Δ + ...
2 3 4 h 12 6
1 F
and D3 =
h H
G Δ − 32 Δ + ...IJK
3
3 4
(2) Newton’s backward difference interpolation formula is

u(u + 1) 2 u(u + 1)(u + 2) 3
y = yn + u ∇yn + ∇ yn + ∇ yn + ... ...(10)
2! 3!
x − xn
where u = ...(11)
h
FG
3u2 + 6u + 2 IJ F I
m
dy 2u + 1 2
du
= ∇yn +
2
∇ yn +
H6 K
∇3yn + ... GH JK ...(12)
o
du 1
.c
= ...(13)
dx h
Now,
dy dy du
=
dx du dx
1
.
LM
2u + 1 2
n
FG
3u 2 + 6u + 2 IJ F
ts I OP
= ∇yn + ∇ yn +
MN H ∇ 3 yn + ...
K GH JK PQ ...(14)
ra
h 2 6
dy
Expression (14) provides us the value of at any x which is not tabulated.
pi
dx
At x = xn, we have u = 0
∴ Putting u = 0 in eqn. (14), we get
as
FG dy IJ 1 1FG 1 1
∇yn + ∇ 2 yn + ∇ 3 yn + ∇ 4 yn + ...
IJ ...(15)
H dx K =
h 2 H 3 4 K
g
x = xn
.c

d2 y d dy du FG IJ
w
dx 2
=
du dx dx H K
1 L F 6u I OP
w
2
+ 18u + 11 4
=
h MN
M∇ y
2
2
n + (u + 1) ∇ 3 yn + GH 12
∇ yn + ... JK PQ ...(16)
w
F d yI
2
1 FG ∇ y 11 4 IJ
GH dx JK
2
x = xn
=
h 2 H
2
n + ∇ 3 yn +
12
∇ yn + ...
K ...(17)
Similarly, we get
F d yI
3
1 FG ∇ y 3 4 IJ
GH dx JK
3
x = xn
=
h 3 H
3
n +
2
∇ yn + ...
K ...(18)
and so on.
www.cgaspirants.com
www.cgaspirants.com
Aliter: We know that

E–1 = 1 – ∇
e–hD = 1 – ∇
1 2 1 3 1 4 FG IJ
∴ – hD = log (1 – ∇) = – ∇ +
2
∇ + ∇ + ∇ + ...
3 4 H K
1 FG
1 2 1 3 1 4 IJ
⇒ D=
h 2 3 H
∇ + ∇ + ∇ + ∇ + ...
4 K
D2 =
FG ∇ + 1 ∇ + 1 ∇ + ...IJ = 1 FG ∇
1 2 3
2
2
+ ∇3 +
11 4 IJ
h H K h H K
Also, 2 2
∇ + ...
2 3 12
1 F
Similarly, D3 =
h H
G ∇ + 2 ∇ + ...IJK and so on.
3 3 4
m
3
(3) Stirling’s central difference interpolation formula is

F I
o
FG
u Δy0 + Δy−1 u2 2
Δ y−1 +
IJ
u(u 2 − 12 ) Δ3 y−1 + Δ3 y−2
GH JK
y = y0 +
H +
K
.c
1! 2 2! 3! 2
u 2 (u 2 − 12 ) 4 u(u 2 − 12 )(u 2 − 2 2 ) Δ5 y−2 + Δ5 y−3 F I + ...
where u=
x−a
+
4!
Δ y−2 +
5!
n 2
ts GH JK ...(19)
...(20)
h
ra

dy Δy0 + Δy−1 3u 2 − 1 F I FΔ y +Δ y I
3 3
= + u Δ2 y−1 + GH JK GH 2 JK −1 −2
pi
du 2 6
F 4u − 2u I Δ y + F 5u
3 4
− 15u + 4 I F Δ y + Δ y I
2 5 5
as
+ GH 4 ! JK GH 4
−2
5! JK GH 2 JK + ... −2 −3
...(21)

g
du 1
= ...(22)
.c
dx h
dy dy du
w
Now, = .
dx du dx
LM 3u2 − 1 F I FΔ y +Δ y I
3 3
w
1 Δy0 + Δy−1
=
h MN
2
+ u Δ2 y−1 +
6 GH JK GH 2 JK −1 −2
w
F 4u − 2u I Δ y + F 5u
3 4
− 15u 2 + 4I F Δ y + Δ y I
5 O 5
+ GH 4 ! JK GH 4
−2
5! JK GH 2 JK + ...PPQ
−2 −3
...(23)
dy
dx
Put x = a, we have u = 0
FG dy IJ 1 LMF Δy + Δ y −1 IJ F
1 Δ3 y−1 + Δ3 y−2 I F
1 Δ5 y−2 + Δ5 y−3 I OP
MNGH GH JK GH JK
0
+ − ...
H dx K x=a
=
h 2
−
6K 2 30 2 PQ
...(24)
www.cgaspirants.com
www.cgaspirants.com

d2 y FG IJ
d dy du
dx 2
=
H K
du dx dx
1 L FΔ y
3
+ Δ3 y−2 I F
6u 2 − 1 I
=
h MN
MΔ y
2
2
−1 +u GH −1
2
+ JK GH
12 JK
Δ4y–2
F 2u − 3u I F Δ y
+G
3 5
+ Δ5 y−3 I OP
H 12 JK GH JK
−2
+ ... ...(25)
2 PQ
F d yI FG Δ y IJ
m
2
1 1 4 1 6
GH dx JK
2
x=a
=
h 2 H
2
−1 −
12
Δ y −2 +
90
Δ y−3 − ...
K ...(26)
o
and so on.
.c
(4) Bessel’s central difference interpolation formula is
Fy
y = GH
0 + y1
2
+ u−
IJ FG
1
K H
2
Δy0 +
2!
n
IJ
u(u − 1) Δ2 y−1 + Δ2 y0
K
2 JK + 3!
F
GH
ts
I u(u − 1) FGH u − 2 IJK Δ y
1
3
−1
ra
F 1I
(u + 1) u(u − 1)(u − 2) G u − J
F
(u + 1) u(u − 1)(u − 2) Δ y + Δ y I 4 4
H 2K Δ y
+ GH 2 JK + −2 −1 5
pi
−2
4! 5!
(u + 2)(u + 1) u(u − 1)(u − 2)(u − 3) F Δ y + Δ y I 6 6
GH 2 JK + ... ...(27) −3 −2
as
+
6!
x−a
g
where u= ...(28)
h
.c

F 1 I
IJ F Δ y I + GG 3u 2
FG − 3u +
JJ
w
2
dy 2u − 1 + Δ2 y0 2
K GH JK H
−1
du
= Δy0 +
2! H 2 3! K Δ3y–1
w
F 4u 3
− 6u 2 − 2u + 2 I FΔ y 4 I F
+ Δ4 y−1 5u 4 − 10u3 + 5u − 1 I
GH JK GH −2
JK GH
+ JK
w
+ Δ5y–2
4! 2 5!
F 6u 5
− 15u 4 − 20u3 + 45u 2 + 8u − 12 I F Δ y 6
+ Δ6 y−2 I + ...
+ GH 6! JK GH −3
2 JK ...(29)

du 1
=
dx h
dy dy du
Now, = .
dx du dx
www.cgaspirants.com
www.cgaspirants.com
LM
F
1 Δy + 2u − 1 I F Δ y + Δ y I + FGG 3u − 3u + 21IJJ 2 2
2
=
h
MM GH JK GH 2 JK H 3 ! K Δ y
0
2!
−1 0
3
–1
MN
+G
F 4u − 6u − 2u + 2 I F Δ y + Δ y I + F 5u − 10u + 5u − 1I Δ y
3 2 4 4 4 3
H 4! JK GH 2 JK GH 5! JK −2 −1 5
–2
F 5
+ G 6u − 15u − 20u + 45u + 8u − 12 J
I F Δ y + Δ y I + ...OP ...(30)
4 3 2 6 6
K GH 2 JK PQ
−3 −2
H 6!
dy
m
dx
Put x = a, we have u = 0
o
.c
FG dy IJ 1 LM
1 Δ2 y−1 + Δ2 y0 1 3 F 1 Δ4 y−2 + Δ4 y−1 I F I
GH JK GH JK
ts
Δy0 − Δ y −1 +
H dx K x=a
=
h 2MN 2
+
12 12 2
F I + ...OP
n
1 5 1 Δ6 y−3 + Δ6 y−2
– Δ y−2 − GH JK PQ ...(31)
ra
120 60 2

pi
d2 y FG IJ = d FG dy IJ du
d dy
dx 2
=
dx H K du H dx K dx
dx
as
1 LF Δ y + Δ y I F 2u − 1I F 6u I FΔ y +Δ y I
2 2 2 4 4
− 6u − 1
= M
h MNH
G 2
2 J +G
−1
K H 2 K J Δ y +G
H
0 3
−1 JK GH 2 JK
12
–2 −1
g
F 4u − 6u + 1I Δ y
.c
3 2
+G
H 24 JK
5
–2
w
F 15u 4
− 30u3 − 30u2 + 45u + 4 I F Δ y + Δ y I 6 O 6
GH JK GH 2 JK + ...PPQ ...(32)−3 −2
w
+
360
w
F d yI
2
1 LMF Δ y 2
+ Δ2 y0 1 I 1 Δ4 y−2 + Δ4 y−1 F I
GH dx JK2
x= a
=
h 2
MNGH
−1
2 2 JK
− Δ3 y−1 −
12 2 GH JK
1 5 1 Δ6 y−3 + Δ6 y−2 F I OP
+
24
Δ y−2 +
90 2
+ ... GH JK PQ ...(33)
and so on.
(5) For unequally spaced values of the argument
(i) Newton’s divided difference formula is
f(x) = f(x0) + (x – x0) f(x0) + (x – x0)(x – x1) 2f(x
0) + (x – x0)(x – x1)
(x – x2) 3f(x 4f(x
0) + (x – x0)(x – x1) (x – x2)(x – x3) 0) + ... ...(34)
www.cgaspirants.com
www.cgaspirants.com
f ′(x) is given by
f ′(x) = f(x0) + {2x – (x0 + x1)} 2f(x + {3x2 – 2x(x0 + x1 + x2)
0)
+ (x0x1 + x1x2 + x2x0)} 3f(x

0) + ... ...(35)
(ii) Lagrange’s interpolation formula is
( x − x1 )( x − x2 ) ... ( x − xn ) ( x − x0 )( x − x2 ) ... ( x − xn )
f(x) = f(x0) + f(x1) + ...
( x0 − x1 )( x0 − x2 ) ... ( x0 − xn ) ( x1 − x0 )( x1 − x2 ) ... ( x1 − xn )
...(36)
f ′(x) can be obtained by differentiating f(x) in eqn. (36).
m
Note 1. Formula (8) can be extended as
F d yI 2
1 FG Δ 11 4 5 5 137 6 7 7 363 8 IJ
GH dx JK =
H
2
− Δ3 + Δ − Δ + Δ − Δ + Δ + ... y0
K
o
2
x= a
h2 12 6 180 10 560
.c
2. Formula (17) can be extended as
F d yI2
GH dx JK 2
x = xn
=
1
h 2
FG ∇
H
2
+ ∇3 +
n 11 4 5 5 137 6
12
∇ + ∇ +
6 180
∇ +
ts
7 7 363 8
10
∇ +
560
∇ + ... yn.
IJ
K
5.11 MAXIMA AND MINIMA OF A TABULATED FUNCTION
ra
dy
pi
Since maxima and minima of y = f(x) can be found by equating to zero and solving the
dx
equation for the argument x, the same method can be used to determine maxima and minima
as
of tabulated function by differentiating the interpolating polynomial.

e.g., if Newton’s forward difference formula is used, we have
g
u(u − 1) 2 u(u − 1)(u − 2) 3

y = y0 + u Δy0 + Δ y0 + Δ y0 + ... ...(1)
2! 3!
.c

w
dy 2u − 1 2 3u2 − 6u + 2 3
= Δ y0 + Δ y0 + Δ y0 + ...
du 2! 3!
w
For maxima or minima,

dy
w
=0
du
2u − 1 2 3u2 − 6u + 2 3
⇒ Δy0 + Δ y0 + Δ y0 + ... = 0 ...(2)
2! 3!
If we terminate LHS series after third differences for convenience, eqn. (2) being a
quadratic in u gives two values of u.
Corresponding to these values, x = a + uh will give the corresponding x at which function
may be maximum or minimum.
d2 y d2 y
For maximum, = (–)ve; For minimum, = (+)ve.
du2 du 2
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
dy
Example 1. Find at x = 0.1 from the following table:
dx
x: 0.1 0.2 0.3 0.4
y: 0.9975 0.9900 0.9776 0.9604.
Sol. Take a = 0.1. The difference table is
x y Δy Δ2 y Δ3 y
m
0.1 0.9975
– 0.0075
o
0.2 0.9900 – 0.0049
.c
– 0.0124 0.0001
0.3 0.9776 – 0.0048
0.4 0.9604
n – 0.0172
ts
ra
Here h = 0.1 and y0 = 0.9975
LM dy OP 1LM 1 1 OP
pi
= Δy0 − Δ2 y0 + Δ3 y0
N dx Q x = 0.1 h N 2 3 Q
as
=
1 LM 1 1
− 0.0075 − (− 0.0049) + (0.0001) = – 0.050167.
OP
0.1 N 2 3 Q
g
Example 2. The table given below reveals the velocity ‘v’ of a body during the time ‘t’
.c
specified. Find its acceleration at t = 1.1. (U.P.T.U. 2014)

t: 1.0 1.1 1.2 1.3 1.4
w
v: 43.1 47.7 52.1 56.4 60.8.

w
Sol. The difference table is

w
t v Δv Δ2 v Δ3 v Δ4 v
1.0 43.1
4.6
1.1 47.7 – 0.2
4.4 0.1
1.2 52.1 – 0.1 0.1
4.3 0.2
1.3 56.4 0.1
4.4
1.4 60.8
www.cgaspirants.com
www.cgaspirants.com
Let a = 1.1,
∴ v0 = 47.7 and h = 0.1
Acceleration at t = 1.1 is given by
LM dv OP =
1 LM 1 1
Δv0 − Δ2 v0 + Δ3 v0 =
1 1 OP 1 LM
4.4 − (− 0.1) + (0.2) = 45.1667
OP
N dt Q t = 1.1 h N 2 3 0.1 2 Q 3 N Q
Hence the required acceleration is 45.1667.
Example 3. Find f ′(1.1) and f ″(1.1) from the following table:
x: 1.0 1.2 1.4 1.6 1.8 2.0
f(x): 0.0 0.1280 0.5540 1.2960 2.4320 4.000.
m
(G.B.T.U. 2010)
Sol. Since we are to find f ′(x) and f ″(x) for non-tabular value of x, we proceed as follows :
o
Newton’s forward difference formula is
.c
u(u − 1) 2 u(u − 1)(u − 2) 3
y = y0 + u Δy0 + Δ y0 + Δ y0
2! 3!
x−a
+
u(u − 1)(u − 2)(u − 3) 4
n
4!
Δ y0 + ... ts
...(1)
where u= ...(2)
ra
h
F I
pi
dy 2u − 1 2 FG
3u2 − 6u + 2 IJ F
2u3 − 9u2 + 11u − 3 I
du
= Δy0 +
2
Δ y0 +
H 6
Δ 3 y0 +
K GH 12 JK
Δ4y0 + ... ...(3) GH JK
as
Differentiating eqn. (2) with respect to x

du 1
...(4)
g
=
dx h
.c
dy dy du
∴ = .
dx du dx
w
1 LM FG
2u − 1 2 IJ
3u 2 − 6u + 2 3 F I
= Δy0 +
MN H Δ y0 +
K GH
Δ y0 JK
w
h 2 6
F 2u 3
− 9u 2 + 11u − 3 4 I OP
w
+ GH 12 JK
Δ y0 + ...
PQ
...(5)
1.1 − 1.0 1 Here a = 1.0

Also, at x = 1.1, u = =
0.2 2 and h = 0.2
www.cgaspirants.com
www.cgaspirants.com
Forward difference table is as follows:
x f(x) = y Δy Δ2 y Δ3 y Δ4 y Δ5 y
1.0 0.0
0.1280
1.2 0.1280 0.298
0.4260 0.018
1.4 0.5540 0.316 0.06
0.7420 0.078 – 0.1
1.6 1.2960 0.394 – 0.04
m
1.1360 0.038
o
1.8 2.4320 0.432
1.5680
.c
2.0 4.000
From eqn. (5),

dy 1 LM 2u − 1 2 FG IJ
n
3u 2 − 6u + 2 3 F I
2u 3 − 9u 2 + 11u − 3 Δ4y
ts F I
= Δy0 +
MN H
Δ y0 +
K Δ y0 + GH JK0 GH JK
ra
dx h 2 6 12
+G
F 5u 4
− 40u 3 + 105u 2 − 100u + 24 5 I OP
H Δ y0 + ... JK ...(6)
pi
120 PQ
At x = 1.1, we get
as
LM R|2 FG 1IJ − 1U|

F I = M0.1280 + S| H 2 K V| (0.298)
f ′(1.1) = G J
dy 1
H dx K 0.2 MN
g
x = 1.1 T 2 W
.c
R|3 FG 1IJ − 6 FG 1IJ + 2U|

2
R|2 FG 1IJ − 9 FG 1IJ + 11 FG 1IJ − 3U|
3 2
S H 2 K H 2 K V| (0.018) + S H 2 K H 2 K H 2 K V (0.06)
w
+| |T |W
T 6 W 12
w
R|5 FG 1IJ − 40 FG 1IJ + 105 FG 1IJ − 100 FG 1IJ + 24U|

4 3 2
OP
+ S H 2K H 2K H 2K H 2 K V (– 0.1) PP
w
|T 120 |W Q
= 0.66724.
Differentiating eqn. (6), with respect to x, we get
d2 y FG IJ
d dy du 1 LM 6u 2 − 18u + 11 4 F I
dx 2
=
H K
du du dx
= 2 Δ2 y0 + (u − 1) Δ3 y0 +
h MN 12
Δ y0 GH JK
F 2u 3
− 12u 2 + 21u − 10 5 I OP
+ GH 12 JK
Δ y0 + ...
PQ
www.cgaspirants.com
www.cgaspirants.com
At x = 1.1, we get
F d yI
2
f ″(1.1) = GH dx JK 2
x = 1.1
= 8.13125.
Example 4. The distance covered by an athlete for the 50 metre race is given in the
following table:
Time (sec): 0 1 2 3 4 5 6
Distance (metre): 0 2.5 8.5 15.5 24.5 36.5 50
Determine the speed of the athlete at t = 5 sec correct to two decimals. (U.P.T.U. 2009)
m
Sol. Here we are to find derivative at t = 5 which is near the end of the table hence we
shall use formula obtained from Newton’s backward difference formula. Backward difference
o
table is as follows:
.c
t s ∇s ∇ 2s ∇ 3s ∇ 4s ∇ 5s ∇ 6s
0 0
1 2.5
2.5
6
n 3.5
– 2.5
ts
2 8.5 1 3.5
ra
7 1 – 3.5
3 15.5 2 0 1
9 1 – 2.5
pi
4 24.5 3 – 2.5
12 – 1.5
as
5 36.5 1.5
13.5
6 50
g
Speed of athlete at t = 5 sec is given by

.c
FG dsIJ LM1 1 1 1 1 OP
w
∇s5 + ∇ 2 s5 + ∇ 3 s5 + ∇ 4 s5 + ∇ 5 s5
H dt K t=5
=
Nh 2 3 4 5 Q
w
1L 1 1 1 1 O
= M12 + (3) + (1) + (0) + (− 3.5)P
1N Q
w
2 3 4 5
= 13.1333 ≈ 13.13 metre/sec.
Example 5. A rod is rotating in a plane. The following table gives the angle θ (in radians)
through which the rod has turned for various values of time t (in seconds)
t: 0 0.2 0.4 0.6 0.8 1.0 1.2
θ: 0 0.12 0.49 1.12 2.02 3.20 4.67.
Calculate the angular velocity and angular acceleration of the rod at t = 0.6 sec.
[G.B.T.U. (C.O.) 2011]
www.cgaspirants.com
www.cgaspirants.com
Sol. Forward difference table is
t θ Δθ Δ2 θ Δ3 θ Δ4 θ
0 0
0.12
0.2 0.12 0.25
0.37 0.01
0.4 0.49 0.26 0
0.63 0.01
0.6 1.12 0.27 0
0.9 0.01
m
0.8 2.02 0.28 0
1.18 0.01
o
1.0 3.20 0.29
1.47
.c
1.2 4.67
Here
∴
a = 0.6
θ0 = 1.12
n
and h = 0.2 ts
Since we are to find derivatives at t = 0.6 sec which is in the middle of the table, hence
ra
we shall use formula obtained from Stirling’s or Bessel’s central difference formula.
Let us choose formula obtained from Bessel’s central difference formula.
pi
Angular velocity at t = 0.6 sec is given by
FG dθ IJ LM I
1 Δ2 θ −1 + Δ2 θ 0 F OP
as
1 1 3
H dt K t = 0.6
=
h
Δθ 0 −
MN 2 2 JK GH
+
12
Δ θ −1
PQ
g
1 L 1 F 0.27 + 0.28 I 1 O
= M0.9 − G
2H
J
K 12 Q
+ (0.01)P = 3.81667 rad./sec.
.c
0.2 N 2
w
Angular acceleration at t = 0.6 sec is given by

F d θI
2
1LMF Δ θ + Δ θ I − 1 Δ θ OP
2 2
GH dt JK G 2 JK 2 PQ
w
−1 0 3
=
h MNH
2 2 −1
t = 0.6
w
1 LF 0.27 + 0.28 I 1 O
=
(0.2) N
MG
H 2
2
J
K − (0.01)P = 6.75 rad./sec .
2 Q
2
Note. In case we choose formula obtained from Stirling’s formula, at t = 0.6 sec.,
angular velocity FG dθ IJ = 1 LMFG Δθ + Δθ IJ − 1 FG Δ θ + Δ θ IJ OP

0 −1
3
−1
3
−2
H dt K h MNH 2 K 6 H 2 K PQ
1 LF 0.9 + 0.63 I 1 F 0.01 + 0.01 I O
= MG 2 JK − 6 GH 2 JK PQ = 3.81667 rad./sec.
0.2 NH
and angular acceleration G

F d θ I = 1 (Δ θ ) = 1 (0.27) = 6.75 rad./sec .
2
H dt JK h
2 2
2 2 −1 2
(0.2)
www.cgaspirants.com
www.cgaspirants.com
dy d2 y
Example 6. From the following table of values of x and y, obtain and for x = 1.2,
dx dx 2
2.2 and 1.6.
x: 1.0 1.2 1.4 1.6 1.8 2.0 2.2
y: 2.7183 3.3201 4.0552 4.9530 6.0496 7.3891 9.0250.
Sol. The forward difference table is
x y Δy Δ2 y Δ3 y Δ4 y Δ5 y Δ6 y
1.0 2.7183
0.6018
1.2 3.3201 0.1333
m
0.7351 0.0294
1.4 4.0552 0.1627 0.0067
o
0.8978 0.0361 0.0013
.c
1.6 4.9530 0.1988 0.0080 0.0001
1.0966 0.0441 0.0014
1.8
2.0
6.0496
7.3891
1.3395
n 0.2429
0.2964
ts
0.0535
0.0094
1.6359
ra
2.2 9.0250
(i) Here a = 1.2
pi
∴ y0 = 3.3201; h = 0.2
LM dy OP LM OP
as
1 1 1 1 1
= 0.7351 − (0.1627) + (0.0361) − (0.008) + (0.0014)
N dx Q x = 1.2 0.2 2 N 3 4 5 Q
g
= 3.3205
LM d y OP
.c
2
=
1 LM0.1627 − 0.0361 + 11 (0.0080) − 5 (0.0014)OP = 3.318
N dx Q2
(0.2) 2
N 12 6 Q
w
x = 1.2
(ii) Here a = 2.2,

w
∴ yn = 9.02 and h = 0.2

LM dy OP LM
w
1 1 1 1
= 1.6359 + (0.2964) + (0.0535) + (0.0094)
N dx Q x = 2.2 0.2 N
2 3 4
+
1 1
(0.0014) + (0.0001) = 9.0229
OP
5 6 Q
LM d y OP
2
=
1 LM
0.2964 + 0.0535 +
11 5
(0.0094) + (0.0014) +
137
(0.0001) = 8.9939.
OP
MN dx PQ2
x = 2.2
0.04 N 12 6 180 Q
(iii) Here a = 1.6
∴ y0 = 4.9530, y–1 = 4.0552, y–2 = 3.3201, y–3 = 2.7183 and h = 0.2
www.cgaspirants.com
www.cgaspirants.com
By using Stirling formula for derivatives, we get

LM dy OP LMFG 1.0966 + 0.8978 IJ − 1 FG 0.0441 + 0.0361IJ
1
N dx Q x = 1.6
=
NH
0.2 2 K 6H 2 K
1 F 0.0014 + 0.0013 I O
+
30 H
G 2
JK PQ = 4.9530
LM d y OP
2
1 L 1 1 O
(0.0001) P = 4.9525.
0.04 MN
and = 0.1988 − (0.0080) +
MN dx PQ
2
x = 1.6
12 90 Q
Example 7. The table below gives the result of an observation. θ is the observed tem-
perature in degrees centigrade of a vessel of cooling water, t is the time in minutes from
m
the beginning of observations:
t: 1 3 5 7 9
o
θ: 85.3 74.5 67.0 60.5 54.3
.c
Find the approximate rate of cooling at t = 3 and 3.5.
Sol. Forward difference table is
t
1 85.3
θ
n
Δθ ts
Δ2 θ Δ3 θ Δ4 θ
ra
– 10.8
3 74.5 3.3
pi
– 7.5 – 2.3
5 67.0 1.0 1.6
– 6.5 – 0.7
as
7 60.5 0.3
– 6.2
g
9 54.3
.c
(i) When t = 3, θ0 = 74.5. Here h = 2

w
dθ
Rate of cooling =
dt
w
FG dθ IJ LM1 1 1 1 OP
Δθ 0 − Δ2θ 0 + Δ3θ 0 − Δ4 θ 0
∴ H dt K =
N Q
w
t=3 h 2 3 4
1L 1 1 O
= M− 7.5 − (1) + (− 0.7)P = – 4.11667°C/min.
2N 2 3 Q
(ii) t = 3.5 is the non-tabular value of t so, we have from Newton’s forward difference
formula,
dθ 1 LM
2u − 1 2 FG
3u 2 − 6u + 2 3IJ F I
=
dt h
Δθ 0 +
2 MN
Δ θ0 +
H 6 K
Δ θ0 GH JK
F 2u 3
− 9u 2 + 11u − 3 4I OP
+ GH 12 JK
Δ θ 0 + ...
PQ ...(1)
www.cgaspirants.com
www.cgaspirants.com
3.5 − 3.0 0.5

At t = 3.5, u= = = 0.25 | Here a = 3.0 and h = 2
2 2
From (1),
FG dθ IJ 1 LM
− 7.5 +
RS
2(0.25) − 1
(1) +
UV R|S
3(0.25) 2 − 6(0.25) + 2
(− 0.7)
U|V OP
H dt K t = 3.5
=
2 MN 2 T W 6 |T |W PQ
= – 3.9151°C/min.
Example 8. Find x for which y is maximum and find this value of y
x: 1.2 1.3 1.4 1.5 1.6
y: 0.9320 0.9636 0.9855 0.9975 0.9996.
m
Sol. The difference table is as follows:
o
.c
1.2 0.9320
0.0316
1.3
1.4
0.9636
0.9855
0.0219
n ts
– 0.0097
– 0.0099
– 0.0002
0.0002
0.0120 0
ra
1.5 0.9975 – 0.0099

0.0021
pi
1.6 0.9996
Let y0 = 0.9320 and a = 1.2

as

g
u(u − 1) 2
y = y0 + u Δy0 + Δ y0 + ...
2
.c
u(u − 1)
= 0.9320 + 0.0316 u + (– 0.0097) | Neglecting higher differences
w
2
dy 2u − 1 FG IJ
= 0.0316 + (− 0.0097)
H K
w
du 2
w
dy
At a maximum, =0
du
FG 1IJ
⇒ 0.0316 = u −
H 2 K
(0.0097) ⇒ u = 3.76
∴ x = a + hu = 1.2 + (0.1) (3.76) = 1.576

To find ymax., we use backward difference formula,
x = xn + hu
⇒ 1.576 = 1.6 + (0.1)u ⇒ u = – 0.24
u(u + 1) 2 u(u + 1)(u + 2) 3
y(1.576) = yn + u ∇yn + ∇ yn + ∇ yn
2 3!
www.cgaspirants.com
www.cgaspirants.com
(− 0.24)(1 − 0.24)
= 0.9996 – (0.24 × 0.0021) + (– 0.0099)
2
= 0.9999988 = 0.9999 nearly
∴ Maximum y = 0.9999 approximately.
Example 9. Derive the formula y0′ = (y–2 – 8y–1 + 8y1 – y2 )/12h
for numerical differentiation at mid-point of a table. (U.P.T.U. 2009)
Sol. Take a = 0 Difference table is
–2 y–2
m
y–1 – y–2
–1 y–1 y0 – 2y–1 + y–2
o
y0 – y– 1 y1 – 3y0 + 3y–1 – y–2
.c
0 y0 y1 – 2y0 + y–1 y2 – 4y1 + 6y0 – 4y–1 + y–2
y1 – y0 y2 – 3y1 + 3y0 – y–1
1
2
y1
y2
y2 – y1
y2 – 2y1 + y0
n ts
ra
By using Stirling’s formula for first derivative, we get
LMF Δy + Δy I − 1 F Δ y + Δ y I + ...OP
1 3 3
MNGH 2 JK 6 GH 2 JK PQ
0 −1 −1 −2
y0′ =
pi
1 LF y − y + y − y I 1 OP
= MG J
as
1 0 0 −1
(y − 3y + 3y − y−1 + y1 − 3 y0 + 3 y−1 − y−2 )
hN H 2 K −
12
2 1 0
Q
1 LF y − y I
= MG JK − 121 ( y − 2 y + 2 y − y )OPQ
g
1 −1
h NH
2 1 −1 −2
2
.c
1
⇒ y0′ = 8 y1 − 8 y−1 − y2 + y−2
w
12h
Example 10. State the three different finite difference approximations to the first
w
derivative f ′(x0) together with the order of their truncation errors. Derive the forward difference
approximation and its leading error term.
w
Sol. (i) Newton’s forward difference approximation is given by

u(u − 1) 2
f(x) = f0 + u Δf0 + Δ f0
2
x − x0 1
where u= and E= u(u – 1) (u – 2) h3 f ′″(ξ)
h 6
We have, f ′(x) =
df du
. =
1 1 LM
Δf0 + (2u − 1) Δ2 f0
OP
du dx h 2 N Q
h2
and |E′(x0)| = |E′(u = 0)| ≤ M3 where M3 = max|f ′′′( x)|
3 x0 ≤ x ≤ x2
www.cgaspirants.com
www.cgaspirants.com
(ii) Newton’s backward difference approximation is given by

1
f(x) = f2 + u ∇f2 + u(u + 1) ∇2 f2
2
x − x2 1
where u = and E= u (u + 1) (u + 2) h3 f ′″(ξ)
h 6
We have, f ′(x) =
1 LM 1
∇f2 + (2u + 1) ∇ 2 f2
OP
h N 2 Q
h2
and |E′(x2)| = |E′ (u = 0)| ≤M3
3
(iii) Central difference approximation is given by
m
u x − x0
f(x) = f0 + (δf1/2 + δf–1/2) where u= .
2 h
o
1 1 1
.c
We have f ′(x) = (δf + δf–1/2) = [(f – f ) + (f0 – f–1)] = (f – f )
2h 1/2 2h 1 0 2h 1 –1
and |E′(x)| ≤
h2
6
M3.
n ts
ASSIGNMENT
ra
1. (i) Find y′(0) and y″(0) from the given table:

x: 0 1 2 3 4 5
pi
y: 4 8 15 7 6 2
(ii) Find the first and second derivatives for the function tabulated below at the point x = 3.0:
as
x: 3 3.2 3.4 3.6 3.8 4.0

y: –14 –10.032 –5.296 0.256 6.672 14
2. (i) From the following table, estimate y′ (1.05):
g
x: 1 1.05 1.10 1.15 1.20 1.25

.c
y: 1.1 1.1347 1.1688 1.1564 1.2344 1.2345 (G.B.T.U. 2013)

(ii) Find the derivative of f(x) at x = 0.4 from the following table:
w
x: 0.1 0.2 0.3 0.4

f(x): 1.10517 1.22140 1.34986 1.49182 (M.T.U. 2013)
w
d d2
3. Find (J0 ) and (J ) at x = 0.1 from the following table:
w
dx dx2 0
x: 0.0 0.1 0.2 0.3 0.4
J0(x): 1 0.9975 0.99 0.9776 0.9604.
4. (i) Find the numerical value of y′(10°) for y = sin x given that:
sin 0° = 0.000, sin 10° = 0.1736, sin 20° = 0.3420, sin 30° = 0.5000, sin 40° = 0.6428.
(ii) Compute y′ (1.1) from the following table:
x: 1.0 1.1 1.2 1.3 1.4 1.5
y (x): 7.989 8.403 8.781 9.129 9.451 9.750 (G.B.T.U. 2012)
5. (i) Find f ′(1.5) and f ″(1.5) from the following table:
x: 0.0 0.5 1.0 1.5 2.0
f(x): 0.3989 0.3521 0.2420 0.1245 0.0540 [U.P.T.U. (MCA) 2006]
www.cgaspirants.com
www.cgaspirants.com
(ii) The population of a certain town is given below. Find the rate of growth of the population in
1961:
Year: 1931 1941 1951 1961 1971
Population: 40.62 60.80 71.95 103.56 132.65 (M.T.U. 2012)
(in lacs)
6. Find first and second derivative of the function tabulated below at x = 0.6
[G.B.T.U. (C.O.) 2010]
x: 0.4 0.5 0.6 0.7 0.8
y: 1.5836 1.7974 2.0442 2.3275 2.6511.
7. Find the values of f ′(5), f ″(5) and f ″(0.5) from the following table:
(G.B.T.U. 2012, U.P.T.U. 2007)
x: 0 1 2 3 4 5
m
f(x): 4930 5026 5122 5217 5312 5407
8. (i) A slider in a machine moves along a fixed straight rod. Its distance x cm along the rod is given
o
below for various values of the time t seconds. Find the velocity of the slider and its accelera-
tion when t = 0.3 second.
.c
t: 0 0.1 0.2 0.3 0.4 0.5 0.6
x: 30.13 31.62 32.87 33.64 33.95 33.81 33.24.
given at various times (sec.)

t: 1 1.1
n
1.2 1.3
ts
(ii) A slider in a machine moves along fixed straight rod. Its distances x (m.) along the rod are
1.4 1.5
ra
x: 16.40 19.01 21.96 25.29 29.03 33.21
Find the velocity of the slider at t = 1.1 sec.
9. A slider in a machine moves along a fixed straight rod. Its distance x(in cm) along the rod is given
pi
at various times t (in secs).

t: 0 0.1 0.2 0.3 0.4 0.5 0.6
as
x: 30.28 31.43 32.98 33.54 33.97 33.48 32.13

dx
Evaluate at t = 0.1 and at t = 0.5.
dt
g
10. Given the following table:

.c
x: 1 1.05 1.1 1.15 1.2 1.25 1.3
f(x) = x : 1 1.0247 1.04881 1.07238 1.09544 1.11803 1.14014

w
Apply the above results to find f ′(1), f ″(1) and f ″′(1).

w
11. (i) Using Newton’s divided difference formula, find f ′(10) from the following data:
x: 3 5 11 27 34
w
f(x): – 13 23 899 17315 35606

(ii) Given the following data, find y′(6):
x: 0 2 3 4 7 8
f(x): 4 26 58 112 466 922
(U.P.T.U. 2008)
(iii) Find f ′(6) from the following table:
x: 0 1 3 4 5 7 9
f(y): 150 108 0 –54 –100 –144 –84
12. (i) From the table below, for what value of x, y is minimum? Also find this value of y.
x: 3 4 5 6 7 8
y: 0.205 0.240 0.259 0.262 0.250 0.224.
www.cgaspirants.com
www.cgaspirants.com
(ii) Find the minimum value of y from the following table:

x: 0.2 0.3 0.4 0.5 0.6 0.7
y: 0.9182 0.8975 0.8873 0.8862 0.8935 0.9086
13. If y = f (x) and yn denotes f(x0 + nh), prove that, if powers of h above h6 be neglected,
FG dy IJ 3 LM 1 1 OP
H dx K x = x0
=
4h
y1 − y−1 − ( y2 − y−2 ) +
N 5 45
( y3 − y−3 )
Q (U.P.T.U. 2006)
14. The following table gives values of pressure P and specific volume V of saturated steam:
P: 105 42.7 25.3 16.7 13
V: 2 4 6 8 10
Find
(a) the rate of change of pressure w.r.t. volume at V = 2
m
(b) the rate of change of volume w.r.t. pressure at P = 105.
15. y is a function of x satisfying the equation xy″ + ay′ + (x – b) y = 0, where a and b are integers. Find
o
the values of constants a and b if y is given by the following table:
x: 0.8 1 1.2 1.4 1.6 1.8 2 2.2
.c
y: 1.73036 1.95532 2.19756 2.45693 2.73309 3.02549 2.3333 3.65563.
16. (a) When does the need of numerical differentiation arise? (U.P.T.U. 2008)
(b) Prove that D =

1
h
F
GH
Δ−
Δ2 Δ3
2
+
3
I
JK ts
− ... , the symbols used have usual meanings.
n
[U.P.T.U. (MCA) 2007]
ra
Answers
pi
1. (i) – 27.9, 117.67 (ii) 18, 18

2. (i) 3.5856 (ii) 1.4913 3. – 0.05, – 0.5
4. (i) 0.9848 (ii) 3.945833
as
5. (i) – 0.2051, 0.194 (ii) 5.167 lacs/year 6. 2.64442, 3.6475

7. 94.15, – 3.4166, 1.6458 8. (i) 5.33 cm/sec, – 45.6 cm/sec2 (ii) 27.7 m/sec.
g
9. 32.44166 cm/sec., – 24.05833 cm/sec. 10. 0.50024, – 0.256, 0.4

.c
11. (i) 232.869 (ii) 125.4744 (iii) – 23

12. (i) 5.6875, 0.2628 (ii) 0.4623
w
14. (a) – 52.4 (b) – 0.01908 15. a = 8, b = 6.

w
5.12 ERRORS IN NUMERICAL DIFFERENTIATION

w
In numerical differentiation, the error in the higher order derivatives occurs due to the reason
that, although the tabulated function and its approximating polynomial would agree at the
set of data points, their slopes at these points may vary considerably. Numerical differentia-
tion is, therefore, an unsatisfactory process and should be used only in rare cases.
The numerical computation of derivatives involves two types of errors: truncation
errors and rounding errors.
The truncation error is caused by replacing the tabulated function by means of an inter-
polating polynomial.
Truncation error in first derivative is
1 Δ3 y−2 + Δ3 y−1
= .
6h 2
www.cgaspirants.com
www.cgaspirants.com
Truncation error in second derivative is

1
= | Δ4 y−2 |.
12 h 2
1 1
The rounding error is ∝ in case of first derivatives while it is ∝ 2 in case of second
h h
derivatives, and so on.
3 ε
Max. Rounding error in first derivative is =
2 h
4ε
Max. Rounding error in second derivative is = 2
m
h
where ε is the maximum error in the value of yi.
o
Example. Assume the table of values given in previous. Example 6 and the function
dy
.c
values are correct to the accuracy given, estimate the errors in at x = 1.6.
dx
Truncation error =
n ts
Sol. Since the values are correct to four decimals, it follows that ε = 0.5 × 10– 4
1 Δ3 y−1 + Δ3 y0
6h 2
ra
FG
1 0.0361 + 0.0441 IJ
=
H K | See difference table in Example 6
pi
1.2 2
= 0.03342
as
3ε 3 × 0.5 × 10 −4
Rounding error = = = 0.00038.
2h 2 × 0.2
g
.c
5.13 NUMERICAL INTEGRATION

w
Given a set of tabulated values of the integrand
z xn
w
f(x), to determine the value of f ( x) dx is

x0
called numerical integration. We subdivide the

w
given interval of integration into a large number

of subintervals of equal width h and replace the
function tabulated at the points of subdivision
by any one of the interpolating polynomials like
Newton-Gregory’s, Stirling’s, Bessel’s over each
of the subintervals and evaluate the integral.
We have several formulae for numerical
integration which we shall derive in the sequel.
www.cgaspirants.com
www.cgaspirants.com
5.14 NEWTON-COTE’S QUADRATURE FORMULA

[M.T.U. 2013, U.P.T.U. (MCA) 2009]
Let I=
za
b
y dx , where y takes the values y0, y1, y2, ......., yn for x = x0, x1, x2, ......, xn.
Let the interval of integration (a, b) be divided into n equal sub-intervals, each of width
b−a
h= so that x0 = a, x1 = x0 + h, x2 = x0 + 2h, ..., xn = x0 + nh = b.
n
∴ I= z x0
x0 + nh
f ( x) dx
m
Since any x is given by x = x0 + rh and dx = hdr
I=h z n
o
∴ f ( x0 + rh) dr
0
z LMN OP
.c
n r(r − 1) 2 r(r − 1)(r − 2) 3
=h y0 + rΔy0 + Δ y0 + Δ y0 + ... dr
0 2! 3! Q
L r Δy 2
n
F
1 r3 r2 2 1 r4 I
ts
[By Newton’s forward interpolation formula]
F I OP n
= h Mry + + GH
− Δ y0 + JK GH
− r 3 + r 2 Δ3 y0 + ... JK
ra
MN 2 0 0
2 3 2 6 4 PQ 0
L n
I = nh M y + Δy
n(2n − 3) 2 n(n − 2) 3 2
+ ...P
O
pi
+ Δ y0 + Δ y0 ...(1)
N 2 0 0
12 24 Q
This is a general quadrature formula and is known as Newton-Cote’s quadrature
as
formula. A number of important deductions viz. Trapezoidal rule, Simpson’s one-third and
three-eighth rules, can be immediately deduced by putting n = 1, 2 and 3 respectively, in
g
formula (1).
.c
5.15 TRAPEZOIDAL RULE (n = 1) [U.P.T.U. MCA (SUM) 2008]

w
Putting n = 1 in formula (1) and taking the curve through (x0, y0) and (x1, y1) as a polynomial
of degree one so that differences of order higher than one vanish, we get
w
z x0 + h
f ( x) dx = h y0 +
FG
H
1 h IJ
K
h
Δy0 = [2 y0 + ( y1 − y0 )] = ( y0 + y1 )
w
x0 2 2 2
Similarly, for the next sub-interval (x0 + h, x0 + 2h), we get
z x0 + 2 h
x0 + h
f ( x) dx =
h
2
( y1 + y2 ) , ......,
z x0 + nh
x0 + ( n − 1) h
f ( x) dx =
h
2
( yn − 1 + yn )
Adding the above integrals, we get
zx0
x 0 + nh
f(x) dx =
h
2
[(y 0 + y n ) + 2(y 1 + y 2 + .. . .. . + y n − 1 )]
which is known as Trapezoidal rule. By increasing the number of subintervals, thereby

making h very small, we can improve the accuracy of the value of the given integral.
www.cgaspirants.com
www.cgaspirants.com
5.16 SIMPSON’S ONE-THIRD RULE (n = 2) (G.B.T.U. 2013)
Putting n = 2 in formula (1) and taking the curve through (x0, y0), (x1, y1) and (x2, y2) as a
polynomial of degree two so that differences of order higher than two vanish, we get
z x0 + 2 h
x0
2h
LM
f ( x) dx = 2h y0 + Δy0 +
N
1 2
6
Δ y0
OP
Q
h
=
[6 y0 + 6( y1 − y0 ) + ( y2 − 2 y1 + y0 )] = ( y0 + 4 y1 + y2 )
6 3
Similarly, z
x0 + 4 h
x0 + 2 h
f ( x) dx =
h
3 2
(y + 4y3 + y4), ...... ,
z
x0 + nh h
f ( x) dx = ( yn − 2 + 4 yn − 1 + yn )
m
x0 + ( n − 2) h 3
o
z x 0 + nh
f(x) dx =
h
[(y0 + yn) + 4(y1 + y3 + ... + yn–1) + 2(y2 + y4 + ... + yn–2)]
.c
x0 3
which is known as Simpson’s one-third rule.

While using this formula, the given interval of integration must be divided
n
into an even number of sub-intervals, since we find the area over two sub-intervals at a
time.
ts
ra
5.17 SIMPSON’S THREE-EIGHTH RULE (n = 3) (U.P.T.U. 2015)

pi
Putting n = 3 in formula (1) and taking the curve through (x0, y0), (x1, y1), (x2, y2) and (x3, y3) as
a polynomial of degree three so that differences of order higher than three vanish, we get
z FG IJ
as
x0 + 3 h 3 3 1
Δy0 + Δ2 y0 + Δ3 y0
x0
f ( x) dx = 3h y0 +
H 2 4 8 K
g
3h
= [8y0 + 12(y1 – y0) + 6(y2 – 2y1 + y0) + (y3 – 3y2 + 3y1 – y0)]
.c
8
3h
w
= [y0 + 3y1 + 3y2 + y3]

8
z 3h
x0 + 6 h
w
Similarly, [y3 + 3y4 + 3y5 + y6], ...

f ( x) dx =
x0 + 3 h 8
z
w
x0 + 6 h 3h
f ( x) dx = [yn–3 + 3yn–2 + 3yn–1 + yn]
x0 + ( n − 3) h 8
zx0
x 0 + nh
f(x) dx =
3h
8
[(y0 + yn) + 3(y1 + y2 + y4 + y5 + ..... + yn–2 + yn–1)
+ 2(y3 + y6 + ...... + yn–3)]
which is known as Simpson’s three-eighth rule.

While using this formula, the given interval of integration must be divided
into sub-intervals whose number n is a multiple of 3.
www.cgaspirants.com
www.cgaspirants.com
5.18 ERRORS IN QUADRATURE FORMULAE
If yp is a polynomial representing the function y = f(x) in the interval [a, b] then error in the
quadrature formulae is given by
E= za
b
f(x) dx – za
b
yp dx ...(1)
5.18.1. Error in Trapezoidal Rule [U.P.T.U. (MCA) 2006, U.P.T.U. MCA (SUM) 2008]
Expanding y = f(x) in the neighbourhood of x = x0 by Taylor’s series, we get
( x − x0 ) 2
y = y0 + (x – x0) y0′ + y0″ + ... ...(2)
m
2!
z z LM y OP
o
x1 x 0 +h ( x − x 0 )2
∴ y dx = + ( x − x 0 ) y0′ + y 0 ″ + ... dx
x0
MN 0
2! PQ
.c
x0
h2
= hy0 +
2!
y0″ +
n h3
y ″′ + ...
3! 0 ts h
...(3)
Now, area of the first trapezium in the interval [x0, x1] = A1 = (y + y1) ...(4)
ra
2 0
Putting x = x0 + h, y = y1 in (2),
pi
h2
y1 = y0 + hy0′ + y ″ + ... ...(5)
2! 0
as
From (4) and (5), we get
LM h2 h2 OP h3
g
h
A1 = y 0 + y 0 + hy 0 ′ + y 0 ″ + ... = hy0 + y0′ + y ″ + ... ...(6)
2 MN 2! 2! PQ2 2! 0
.c
Subtracting eqn. (6) from eqn. (3) gives the error in [x0, x1],
w
z x1
y dx – A1 =
FG 1 − 1 IJ h
H 3 ! 2 2 !K
3 y0″ + ... ≈ –
h3
y ″ | Neglecting other terms
w
x0 12 0
w
h3 − h3
Similarly, the error in [x1, x2] is – y1″ and in [xn–1, xn] is y″ .
12 12 n–1
Hence the total error is
− h3
E=
(y0″ + y1″ + ··· + y″n–1)
12
Let y″(ξ), a < ξ < b be the maximum of | y0″ |, | y1″ |,..., | y″n–1| then, we have
− nh 3 (b − a) 2
E< y″(ξ) = – h y″(ξ) |∵ b – a = nh
12 12
Hence the error in the Trapezoidal rule is of order h2.
www.cgaspirants.com
www.cgaspirants.com
5.18.2. Error in Simpson’s 1/3rd Rule. (U.P.T.U. 2008)
Integrating eqn. (2) w.r.t. x between the limits x0 and x2.
zx0
x2
y dx = zx0
x 0 + 2h LM y
MN 0 + ( x − x 0 ) y0′ +
( x − x 0 )2
2!
OP
y 0 ″ + ... dx
PQ
8h 3 16h 4 32h 5
= 2hy0 + 2h2y0′ + y 0″ + y 0″′ + y0(iv) + ... ...(7)
3! 4! 5!
h
Now, A1 =
(y + 4y1 + y2) ...(8)
3 0
where A1 is the area of the curve in the interval [x0, x2].
m
Putting x = x0 + h, y = y1 in (2), we get
o
h2 h3
y1 = y0 + hy0′ + y 0″ + y ″′ + ... ...(9)
3! 0
.c
2!
Putting x = x0 + 2h, y = y2 in (2), we get
y2 = y0 + 2h y0′ +
n 4h 2
2!
y 0″ +
8h 3
3! 0
ts
y ″′ + ... ...(10)
Substituting eqns. (9) and (10) in eqn. (8), we get

ra
4h 3 2h 4 5h 5 (iv)
y 0″ +
A1 = 2h y0 + 2h2 y0′ + y 0″′ + y + ... ...(11)
3 3 18 0
pi
Now, the error in interval [x0, x2] is given by
z x2 FG 4 − 5 IJ h
as
x0
y dx – A1 = H 15 18 K 5 y0(iv) + ...
h 5 (iv)
g
≈– y | Neglecting terms of order h6, h7, ...

90 0
.c
h 5 (iv)
Similarly, the principal part of error in interval [x2, x4] is = – y and so on.
w
90 2
Hence total principal error is
w
h5
E=– [y (iv) + y2(iv) + ... + y2(iv)
(n–1)]
w
90 0
Let y(iv) (ξ) be the maximum of | y0(iv) |, | y2(iv) |,..., | y2(iv)
n–1 |.
h 5 (iv) − (b − a) h 4 (iv)
Then, we have, E < – y (ξ) = y (ξ)
90 180
Hence, the error in the Simpson’s (1/3)rd rule is of order h4.
3h5 (iv)
Note. Similarly, the principal part of the error for Simpson’s (3/8)th rule is – y in the interval
80
[x0, x3].
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
Example 1. Evaluate
z0.6
2
y dx , where y is given by the following table:
x: 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

y: 1.23 1.58 2.03 4.32 6.25 8.36 10.23 12.45.
Sol. Here the no. of sub-intervals is 7 which is neither even nor a multiple of 3. Also this
number is neither a multiple of 4 nor a multiple of 6 hence using Trapezoidal rule, we get
z 2
y dx =
h
[(y0 + y7) + 2(y1 + y2 + y3 + y4 + y5 + y6)]
m
0.6 2
0.2
= [(1.23 + 12.45) + 2(1.58 + 2.03 + 4.32 + 6.25 + 8.36 + 10.23)]
o
2
| Here h = 0.2
.c
= 7.922.
from x = 7.47 to x = 7.52.

x: 7.47 7.48
n
7.49 7.50
ts
Example 2. Find, from the following table, the area bounded by the curve and the x-axis
7.51 7.52
ra
f(x): 1.93 1.95 1.98 2.01 2.03 2.06.

Sol. We know that
pi
Area = z 7.52
7.47
f ( x) dx with h = 0.01, the trapezoidal rule gives,
as
0.01
Area = [(1.93 + 2.06) + 2(1.95 + 1.98 + 2.01 + 2.03)] = 0.09965.
2
g
z 0.3
.c
Example 3. Use Simpson’s rule for evaluating f(x) dx from the table given below:
−0.6
w
x: – 0.6 – 0.5 – 0.4 – 0.3 – 0.2 – 0.1 0 0.1 0.2 0.3

f(x): 4 2 5 3 –2 1 6 4 2 8
w
Sol. Since the no. of sub-intervals is 9(a multiple of 3) hence we will use Simpson’s 3/8th
rule here.
w
∴
z 0.3
−0.6
3(0.1)
[(4 + 8) + 3{2 + 5 + (– 2) + 1 + 4 + 2} + 2(3 + 6)] = 2.475.
f ( x) dx =
8
Example 4. A river is 80 m wide. The depth ‘y’ of the river at a distance ‘x’ from one bank
is given by the following table:
x: 0 10 20 30 40 50 60 70 80
y: 0 4 7 9 12 15 14 8 3
Find the approximate area of cross section of the river using Simpson’s 1/3rd rule.
Sol. The required area of the cross section of the river = z 0

80
y dx ...(1)
www.cgaspirants.com
www.cgaspirants.com
Here no. of sub-intervals is 8. By Simpson’s 1/3rd rule,
z0
80 h
[(y + y8) + 4(y1 + y3 + y5 + y7) + 2(y2 + y4 + y6)]
y dx =
3 0
10
= [(0 + 3) + 4(4 + 9 + 15 + 8) + 2(7 + 12 + 14)] = 710
3
Hence the required area of the cross section of the river = 710 sq. m.
Example 5. The speed, v metres per second, of a car, t seconds after it starts, is shown in
the following table:
t 0 12 24 36 48 60 72 84 96 108 120
v 0 3.60 10.08 18.90 21.60 18.54 10.26 5.40 4.50 5.40 9.00
m
FG 1 IJ rd
o
Using Simpson’s rule, find the distance travelled by the car in 2 minutes.
H3K
.c
Sol. If s metres is the distance covered in t seconds, then
ds
dt
=v
n ts
LMsOP t = 120
z 120
ra
= v dt
∴
NQ t =0 0
Since the number of sub-intervals is 10 (even), hence, by using Simpson’s 1/3rd rule,
pi
z 120
v dt =
h
[(v0 + v10) + 4(v1 + v3 + v5 + v7 + v9) + 2(v2 + v4 + v6 + v8)]
as
0 3
12
=
[(0 + 9) + 4(3.6 + 18.9 + 18.54 + 5.4 + 5.4)
g
3
+ 2(10.08 + 21.6 + 10.26 + 4.5)]
.c
= 1236.96 metres.
w
Hence, the distance travelled by car in 2 minutes is 1236.96 metres.

Example 6. A train is moving at the speed of 30 m/sec. Suddenly brakes are applied.
w
The speed of the train per second after t seconds is given by

Time (t): 0 5 10 15 20 25 30 35 40 45
w
Speed (v): 30 24 19 16 13 11 10 8 7 5
Apply Simpson’s three-eighth rule to determine the distance moved by the train in
45 seconds.
Sol. If s metres is the distance covered in t seconds, then
ds
=v
dt
⇒
LMsOP
NQ
t = 45
t =0
= z 45
0
v dt ...(1)
FG 3 IJ th
Since the no. of sub-intervals is 9 (a multiple of 3) hence by using Simpson’s
H 8K rule,
www.cgaspirants.com
www.cgaspirants.com
z
0
45
v dt =
3h
8
15
[(v0 + v9) + 3(v1 + v2 + v4 + v5 + v7 + v8) + 2(v3 + v6)]
= [(30 + 5) + 3(24 + 19 + 13 + 11 + 8 + 7) + 2(16 + 10)]

8
= 624.375 metres.
Hence the distance moved by the train in 45 seconds is 624.375 metres.
Example 7. Evaluate z 1
0
dx
1 + x2
using
1 1
(i) Simpson’s rule taking h = [U.P.T.U. (MCA) 2007]
m
3 4
3 1
(ii) Simpson’s rule taking h = [G.B.T.U. 2010, U.P.T.U. (MCA) 2006]
o
8 6
.c
Hence compute an approximate value of π in each case.
1 1 2 3
Sol. (i) The values of f(x) = n 1 + x2
1
4 4 4
1
ts
at x = 0, , , , 1 are given below:
3
x: 0 1
ra
4 2 4
16
f(x): 1 0.8 0.64 0.5
17
pi
y0 y1 y2 y3 y4
as
1
By Simpson’s rule,
3
z
g
1 dx h
2
= [( y0 + y4) + 4( y1 + y3) + 2y2]
.c
0 1+ x 3
=
1 LM 16 RS
+ 0.64 + 2(0.8) = 0.785392156
UV OP
w
(1 + 0.5) + 4
12 N 17 T W Q
z L O
w
1
dx π
= M tan xP
1
–1
Also = tan –1 1 =
0 1+ x 2
MN PQ 4
w
0
π~
∴ – 0.785392156 ⇒ π ~
– 3.1415686
4
1 1 2 3 4 5
(ii) The values of f(x) = at x = 0, , , , , , 1 are given below:
1 + x2 6 6 6 6 6
1 2 3 4 5
x: 0 1
6 6 6 6 6
36 9 4 9 36 1
f(x): 1
37 10 5 13 61 2
y0 y1 y2 y3 y4 y5 y6
www.cgaspirants.com
www.cgaspirants.com
3
By Simpson’s rule,
8
z
1 dx
0 1+ x 2
=
3h
8
[(y0 + y6) + 3(y1 + y2 + y4 + y5) + 2y3]
FG 1IJ
3
H 6 K LFG 1 + 1IJ + 3 RS 36 + 9 + 9 + 36 UV + 2 FG 4 IJ O = 0.785395862
8 MNH 2K T 37 10 13 61 W H 5 K PQ
=
Also, z 1
0
dx
1+ x 2
=
π
4
π
m
∴ = 0.785395862 ⇒ π = 3.141583
4
6 dx
z
o
Example 8. Evaluate by using
0 1 + x2
.c
(i) Simpson’s one-third rule [M.T.U. (MCA) 2012]
(ii) Simpson’s three-eighth rule
(iii) Trapezoidal rule.
n
Sol. Divide the interval (0, 6) into six parts each of width h = 1.
ts
ra
1
The values of f(x) = are given below:
1 + x2
x: 0 1 2 3 4 5 6
pi
1 1 1
f(x): 1 0.5 0.2 0.1
as
17 26 37
y0 y1 y2 y3 y4 y5 y6
(i) By Simpson’s one-third rule,
g
z 6 dx h
.c
2
= [(y0 + y6) + 4(y1 + y3 + y5) + 2(y2 + y4)]
0 1+ x 3
w
1 LMFG 1 + 1 IJ + 4 FG 0.5 + 0.1 + 1 IJ + 2 FG 0.2 + 1 IJ OP = 1.366173413.

=
3 NH 37 K H 26 K H 17 K Q
w
(ii) By Simpson’s three-eighth rule,

w
z
0
6 dx
1+ x 2
=
3h
8
[(y0 + y6) + 3(y1 + y2 + y4 + y5) + 2y3]
3 LMFG 1 + 1 IJ + 3 FG 0.5 + 0.2 + 1 + 1 IJ + 2(0.1)OP = 1.357080836.

=
8 NH 37 K H 17 26 K Q
(iii) By Trapezoidal rule,
z 6
0
dx
1+ x 2
h
= [(y0 + y6) + 2(y1 + y2 + y3 + y4 + y5)]
2
1 LMFG 1 + 1 IJ + 2 FG 0.5 + 0.2 + 0.1 + 1 + 1 IJ OP = 1.410798581.

=
2 NH 37 K H 17 26 K Q
www.cgaspirants.com
www.cgaspirants.com
Example 9. Evaluate
Hence find loge 2 approximately.

dx
1+ x z 1
by dividing the interval of integration into 8 equal parts.
0
[U.P.T.U. (MCA) 2008]

Sol. Since the interval of integration is divided into an even number of sub-intervals,
we shall use Simpson’s one-third rule.
1
Here, y= = f(x)
1+ x
1 1 1 8 FG IJ FG 2 IJ = 4
y0 = f(0) =
1+ 0
= 1, y1 = f
8
=
1+
= ,
1 9 H K y2 = f
H 8K 5
8
m
FG IJ
3 8 4 2 FG IJ FG 5 IJ = 8
y3 = f
8H K
=
11
, y4 = f
8
= ,
3 H K y5 = f
H 8 K 13
o
F 6I 4
=fG J= ,
F 7I 8
=fG J=
1
.c
y6
H 8K 7 y7
H 8 K 15 and y8 = f(1) =
2
Hence the table of values is
x: 0
1
8
2
8
n 3
8
4
8
ts 5
8
6
8
7
8
1
ra
8 4 8 2 8 4 8 1
y: 1
9 5 11 3 13 7 15 2
pi
y0 y1 y2 y3 y4 y5 y6 y7 y8
By Simpson’s 1/3rd rule,
as
z
0
1 dx h
= [(y0 + y8) + 4(y1 + y3 + y5 + y7) + 2(y2 + y4 + y6)]
1+ x 3
g
1 LMFG 1 + 1IJ + 4 FG 8 + 8 + 8 + 8 IJ + 2 FG 4 + 2 + 4 IJ OP
.c
=
24 NH 2 K H 9 11 13 15 K H 5 3 7 K Q | Here h = 1/8
w
= 0.69315453
w
z LM OP 1
1 dx
Since, = log e (1 + x) = loge 2
MN PQ
w
0 1+ x
0
∴ loge 2 = 0.69315453.
Example 10. Using Simpson’s 3/8th rule on integration, evaluate
z 6
0
1
1+ x
dx [G.B.T.U. 2009; G.B.T.U. (M.Tech.) 2011]
Sol. The table of values is as follows: (Here h = 1)

x: 0 1 2 3 4 5 6
1 1 1 1 1 1 1
f (x) = : 1
1+ x 2 3 4 5 6 7
y0 y1 y2 y3 y4 y5 y6
www.cgaspirants.com
www.cgaspirants.com
By Simpson’s 3/8th rule,
z0
6 1
1+ x
dx =
3h
8
[(y0 + y6) + 3 (y1 + y2 + y4 + y5) + 2y3]
3 LMFG 1 + 1IJ + 3 FG 1 + 1 + 1 + 1IJ + 2 FG 1IJ OP = 1.96607.

=
8 NH 7 K H 2 3 5 6 K H 4 K Q
Example 11. Find z 6
0
ex
1+ x
dx approximately using Simpson’s 3/8th rule on integration.
[U.P.T.U. (MCA) 2009, U.P.T.U. 2006, 2014]
m
Sol. Divide the given integral of integration into 6 equal sub-intervals, the arguments
are 0, 1, 2, 3, 4, 5, 6; h = 1.
o
ex
f(x) = ; y0 = f(0) = 1
.c
1+ x
y1 = f(1) =
e
2
e4
,
n y2 = f(2) =
e2
3
e5
,
ts y3 = f(3) =
e3
4
e6
,
y4 = f(4) = , y5 = f(5) = , y6 = f(6) =

ra
5 6 7
The table is as below:
pi
x: 0 1 2 3 4 5 6
e e2 e3 e4 e5 e6
as
y: 1
2 3 4 5 6 7
y0 y1 y2 y3 y4 y5 y6
g
Applying Simpson’s three-eighth rule, we have

.c
z 6 ex
dx =
3h
[(y0 + y6) + 3(y1 + y2 + y4 + y5) + 2y3]
w
0 1+ x 8
LMF 1 + e I + 3 F e + e + e I OP
w
6 2 4
3 e5 e3
MNGH 7 JK GH 2 3 5 JK
= + +2
8 6 4 PQ
w
3
= [(1 + 57.6327) + 3(1.3591 + 2.463 + 10.9196 + 24.7355 + 2(5.0214)]
8
= 70.1652.
Note. It is not possible to evaluate

ex
0 1+ x
comes to our rescue in such situations.
z 6
dx by using usual calculus method. Numerical integration
www.cgaspirants.com
www.cgaspirants.com
Example 12. Evaluate

Sol. The table of values is:
z
1
2 −
e
1
2
x
dx using four intervals.
x: 1 1.25 1.5 1.75 2

y=e :–x/2 0.60653 0.53526 0.47237 0.41686 0.36788
y0 y1 y2 y3 y4
rd
Since we have four (even) sub-intervals here, we will use Simpson’s 1/3 rule.
∴
z1
2 −
e
1
2
x
dx =
h
3
[(y0 + y4) + 4(y1 + y3) + 2y2]
0.25
= [(0.60653 + 0.36788) + 4(0.53526) + 0.41686) + 2(0.47237)]
m
3
= 0.4773025.
o
1 xp
Example 13. Compute Ip = dx for p = 0, 1.
x 3 + 10
.c
0
Use Trapezoidal rule with number of points 3, 5 and 9.
Sol. For p = 0
I0 =
1 1
0 x 3 + 10
dx z
n ts
1− 0 1
ra
(i) Dividing the interval (0, 1) into 2 equal parts, each of width h = = , the values
2 2
1
pi
of f(x) = 3 are given below:

x + 10
1
as
x: 0 1
2
1 8 1
g
f(x):
10 81 11
.c
By Trapezoidal rule,
h 1 LMFG 1 + 1 IJ + 16 OP = 0.0971099
w
I0 =
2
[(y0 + y2) + 2y1] =
4 NH 10 11K 81 Q
w
1− 0 1
(ii) Dividing the interval (0, 1) into 4 equal parts, each of width h = = , the values
4 4
w
1
of f(x) = 3 are given below:
x + 10
1 1 3
x: 0 1
4 2 4
1 64 8 64 1
f(x):
10 641 81 667 11
h
I0 = [(y0 + y4) + 2(y1 + y2 + y3)]
2
1 1 1 64LMFG 8 64 IJ FG IJ OP = 0.0975039
=
8 10 11
+ +2
NH
+
641 81 667
+
K H KQ
www.cgaspirants.com
www.cgaspirants.com
1
(iii) Dividing the interval (0, 1) into 8 equal parts, each of width h = , the values of
8
1
f(x) = 3 are given below:
x + 10
1 1 3 1 5 3 7
x: 0 1
8 4 8 2 8 4 8
1 512 64 512 8 512 64 512 1
f(x):
10 5121 641 5147 81 5245 667 5463 11
m
h
I0 = [(y0 + y8) + 2 (y1 + y2 + y3 + y4 + y5 + y6 + y7)]
2
o
1 LMFG 1 + 1 IJ + 2 FG 512 + 64 + 512 + 8 + 512 + 64 + 512 IJ OP
=
NH 10 11K H 5121 641 5147 81 5245 667 5463 K Q
.c
16
= 0.0976012
Repeating the above process for p = 1 so that
n ts
z 1 x
ra
I1 = dx
0 x 3 + 10
pi
and dividing the interval (0, 1) into 2, 4 and 8 equal parts respectively, we get I1 as 0.0480733,
0.0481145 and 0.0481164 respectively.
as
Example 14. A solid of revolution is formed by rotating about x-axis, the lines x = 0 and
x = 1 and a curve through the points with the following co-ordinates.
x: 0 0.25 0.5 0.75 1
g
y: 1 0.9896 0.9589 0.9089 0.8415

.c
Estimate the volume of the solid formed using Simpson’s rule.

Sol. If V is the volume of the solid formed then we know that
w
V= π z 1
y 2 dx
w
Hence we need the values of y2 and these are tabulated below correct to four decimal
w
places
x 0 0.25 0.5 0.75 1
y2 1 0.9793 0.9195 0.8261 0.7081
With h = 0.25, Simpson’s rule gives

(0.25)
V= π [(1 + 0.7081) + 4(0.9793 + 0.8261) + 2(0.9195)] = 2.8192.
3
www.cgaspirants.com
www.cgaspirants.com
Example 15. A tank is discharging water through an orifice at a depth of x metre below
the surface of the water whose area is A m2. Following are the values of x for the corresponding
values of A.
A: 1.257 1.39 1.52 1.65 1.809 1.962 2.123 2.295 2.462 2.650 2.827
x: 1.5 1.65 1.8 1.95 2.1 2.25 2.4 2.55 2.7 2.85 3
Using the formula (0.018) T =

x
of the water to drop from 3.0 m to 1.5 m above the orifice.
z1.5
3.0 A
dx , calculate T, the time (in seconds) for the level
Sol. Here h = 0.15

A
m
The table of values of x and the corresponding values of is
x
o
x 1.5 1.65 1.8 1.95 2.1 2.25 2.4 2.55 2.7 2.85 3
.c
A
y= 1.025 1.081 1.132 1.182 1.249 1.308 1.375 1.438 1.498 1.571 1.632
x
Using Simpson’s 1/3rd rule, we get ts

z
n
3 0.15A
dx =
[(1.025 + 1.632) + 4(1.081 + 1.182 + 1.308 + 1.438 + 1.571)
ra
1.5 x 3
+ 2(1.132 + 1.249 + 1.375 + 1.498)]
= 1.9743
pi
Using the formula
z 3 A
as
(0.018)T = dx
x 1.5
We get 0.018T = 1.9743 ⇒ T = 110 sec. (approximately).

g
.c
ASSIGNMENT
w
1. (i) Evaluate
1
z 2
dx by Simpson’s 1/3rd rule with four strips and determine the error by direct
w
1 x
integration. [G.B.T.U. (MCA) 2010]
z
w
1
(ii) Evaluate the integral e x + 1 dx using Simpson’s 1/3 rule by dividing the interval of
0
integration into 8 equal parts. (G.B.T.U. 2013, 2011)
2. Compute z 0
π/2
sin x dx using Simpson’s three-eighth rule of integration, taking h =
π
18
.
(U.P.T.U. 2007)
3. Evaluate
z 4
5.2
log e x dx by Simpson’s 3/8th rule. [G.B.T.U. (C.O.) 2011]
4. (i) Find the value of
intervals.
z
0
π/2
1 − 0.162 sin 2 x dx using Simpson’s one-third rule taking 6 sub-
(U.P.T.U. 2007)
www.cgaspirants.com
www.cgaspirants.com
(ii) Find the approximate value of the following integral using Simpson’s rule by dividing the
interval into six equal parts: z0

π
1 + 3 cos2 x dx . (M.T.U. 2012)
sin ( x)
5. (i) Use Simpson’s 1/3 rule to evaluate the integral of f(x) = with respect to x between
x
x = 0 and x = 1 with a value of h = 0.25. [G.B.T.U. MCA (SUM) 2010]
(ii) Evaluate
z 0
8
x sec x dx using eight intervals by Trapezoidal rule.
(iii) State the need and scope of numerical integration. Use the Trapezoidal rule to estimate the
(U.P.T.U. 2009)
integral
z 2 2
e x dx taking the number of intervals 10. [G.B.T.U. (C.O.) 2010, U.P.T.U. 2008]
m
0
6. (a) Evaluate using Trapezoidal rule
z π
z 2 t dt
o
(i) t sin t dt (ii)
0 −2 5 + 2t
z
.c
2π
(b) Evaluate the integral e− t sin 10 t dt using
0
(i) Simpson’s 3/8 rule.
(ii) Simpson’s rule with eight intervals. ts (M.T.U. 2012)
(U.P.T.U. 2014)
z
n
4
7. (i) Evaluate ex dx by Simpson’s rule, given that e = 2.72, e2 = 7.39, e3 = 20.09, e4 = 54.6 and
ra
0
compare it with the actual value.
z 3
pi
(ii) Using Simpson’s 1/3 rule, evaluate (2 x − x2 ) dx by taking n = 6. (G.B.T.U. 2013)

0
8. (i) The velocities of a car running on a straight road at intervals of 2 minutes are given below:
as
Time (in minutes: 0 2 4 6 8 10 12

Velocity (in km/hr: 0 22 30 27 18 7 0
g
Apply Simpson’s rule to find the distance covered by the car. (G.B.T.U. 2011)
(ii) The velocity v of a particle at distance s from a point on its path is given by the table below:
.c
s (in meter): 0 10 20 30 40 50 60
v (m/sec): 47 58 64 65 61 52 38
w
1
Estimate the time taken to travel 60 meters by using Simpson’s rule. Compare the result
w
3
with Simpson’s 3/8 rule. (U.P.T.U. 2007, 2015 ; G.B.T.U. (C.O.) 2011]
w
(iii) The velocity ‘v’ of a particle at distance ‘s’ from a point on its linear path is given in the
following table:
s (m): 0 2.5 5 7.5 10 12.5 15 17.5 20
v (m/sec): 16 19 21 22 20 17 13 11 9
Apply Simpson’s rule to estimate the time taken by the particle to traverse the distance of 20
meters. (G.B.T.U. 2011)
9. The velocity of a train which starts from rest is given by the following table, time being reckoned
in minutes from the start and speed in kilometres per hour:
Minutes: 0 2 4 6 8 10 12 14 16 18 20
Speed (km/hr): 0 10 18 25 29 32 20 11 5 2 0
Estimate the total distance in 20 minutes.

LMHint. Here step-size h = 2 OP
N 60 Q
www.cgaspirants.com
www.cgaspirants.com
10. A rocket is launched from the ground. Its acceleration is registered during the first 80 seconds
and is given in the table below. Using Simpson’s 1/3rd rule, find the velocity of the rocket at
t = 80 seconds. (G.B.T.U. 2012)
t(sec): 0 10 20 30 40 50 60 70 80
f(cm/sec 2 ): 30 31.63 33.34 35.47 37.75 40.33 43.25 46.69 50.67.
11. (i) Find an approximate value of loge 5 by calculating to 4 decimal places, by Simpson’s 1/3rd
rule, z0
5 dx
4x + 5
dividing the range into 10 equal parts.
(ii) Find the value of loge 2 from

1 + x3 z 1. x2
dx using Simpson’s 1/3rd rule by dividing the range of
0
m
integration into four equal parts. Also find the error.
12. In an experiment, a quantity G was measured as follows:
o
G(20) = 95.9, G(21) = 96.85, G(22) = 97.77, G(23) = 98.68, G(24) = 99.56, G(25) = 100.41,
G(26) = 101.24.
.c
Compute
z 26
G(x) dx by Simpson’s rule.
13.
20
Using the data of the following table, compute the integral

n ts z0.5
1.1
xy dx by Simpson’s rule:
ra
x: 0.5 0.6 0.7 0.8 0.9 1.0 1.1
y: 0.4804 0.5669 0.6490 0.7262 0.7985 0.8658 0.9281
z
pi
1
14. Apply Simpson’s 1/3rd rule to evaluate the integral I = e x dx by choosing step size h = 0.1
0
as
Show that this step size is sufficient to obtain the result correct to five decimal places.
15. (i) Use Simpson’s rule dividing the range into ten equal parts to show that:
z
g
1 log (1 + x 2 )
dx = 0.173
.c
0 1 + x2
z
w
2 dx
(ii) Evaluate to three decimals, dividing the range of integration into 8 equal parts.
0 x2 + x + 1
w
(U.P.T.U. 2007)
z
w
(iii) Compute I =
1.5 2
e− x dx using Simpson’s FG 1IJ rd rule with four subdivisions.
0.2 H 3K
(iv) Using Simpson’s 1/3rd formula, evaluate
z
0
1
3
x +x+1
dx
choose with steplength 0.25.
(U.P.T.U. 2009)
16. (i) Using 3/8th Simpson’s rule, Evaluate:

z0
6
1 + x4
dx
.
(ii) Evaluate
z 9
0
1
1 + x3
dx by using Simpson’s three-eighth rule. [G.B.T.U. (M.Tech.) 2010]
www.cgaspirants.com
www.cgaspirants.com
(iii) State Simpson’s three-eighth rule. Using this rule, evaluate the following integral
z0
6 x
1 + x5
dx (U.P.T.U. 2015)
17. A solid of revolution is formed by revolving the area y = f(x) from x = 1 to x = 3 about the x-axis.
The following table gives relation between x and y for this curve:
x: 1 1.5 2 2.5 3
y: 1 0.9896 0.9589 0.9089 0.8415
Using Simpson’s one-third rule, estimate the volume of the solid formed. (M.T.U. 2013)
18. (i) Find the distance between two stations from the following data consisting of the speeds v(t) of
an electric train at various times t after leaving one station until it stops at the next station.
m
Apply Simpson’s rule:
v (miles/hr): 0 13 33 39.5 40 40 36 15 0
t (min): 0 0.5 1 1.5 2 2.5 3 3.5 4
o
LMHint. Here h = 1 OP
.c
(G.B.T.U. 2012)
N 120 Q
ts
(ii) The speed of a train at various times after leaving one station until it stops at another station
are given in the following table:
Speed (in mph): 0 13 33 39.5 40 40 36 15 0
n
Time (in minutes): 0 0.5 1 1.5 2 2.5 3 3.25 3.5
ra
1
Find the distance between the two stations using trapezoidal rule, Simpson’s rule and
3
pi
3
Simpson’s rule. (M.T.U. 2013)
8
as
LMHint. Since time interval is not same, hence, OP

N z z z
MMs = v dt = v dt + v dt (taking h = 0.5 and 0.25)PP
3.5 3 3.5
Q
g
0 0 3
.c
19. A curve is drawn to pass through the points given by the following table:
x: 1 1.5 2 2.5 3 3.5 4
w
y: 2 2.4 2.7 2.8 3 2.6 2.1

w
Find
(i) Centre of gravity of the area. (ii) Volume of the solid of revolution.
w
(iii) The area bounded by the curve, the x-axis and lines x = 1, x = 4.
20. (i) Obtain the global truncation error term of trapezoidal method of integration.
(ii) Derive an expression for error estimation in Simpson’s one-third rule. (U.P.T.U. 2008)
(iii) Derive Simpson’s 3/8th rule. [U.P.T.U. (MCA) 2009; U.P.T.U. 2009]
Answers
1. (i) 0.69325 ; 0.0001 (ii) 4.67078 2. 0.974353 3. 1.8278
4. (i) 1.505103 (ii) 5.4059
5. (i) 0.946058 (ii) – 6.435936 (iii) 17.1702
6. (a) (i) 3.14 (ii) – 0.747 (b) (i) – 0.24245 (ii) 0.39466
www.cgaspirants.com
www.cgaspirants.com
7. (i) 53.87, 53.60 (ii) 0

5
8. (i) 3 km (ii) 1.063521 sec, 1.0643752 sec (iii) 1.26164 sec
9
9. 5.156 km 10. 30.87 m/sec 11. (i) 1.61 (ii) 0.693255; 0.0001078
12. 591.85333 13. 0.358487 14. 1.718282782
15. (ii) 0.824 (iii) 0.658596 (iv) 0.6305
16. (i) 1.019286497 (ii) 1.124955 (iii) 0.646382112
17. 5.638478 18. (i) 1.8 miles (ii) 1.666 miles, 1.666 miles, 1.6671875 miles
19. (i) (2.53, 1.31) (ii) 64.07 (iii) 7.7833
m
5.19 INTRODUCTION TO ORDINARY DIFFERENTIAL EQUATIONS
A physical situation that concerns with the rate of change of one quantity with respect to
o
another gives rise to a differential equation.
.c
Consider the first order ordinary differential equation
dy
with the initial condition

dx
= f (x, y)
y(x0) = y0
n ts ...(2)
Many analytical techniques exist for solving such equations. But these methods can be
...(1)
ra
applied to solve only a selected class of differential equations.

However, a majority of differential equations appearing in physical problems cannot be
pi
solved analytically. Thus it becomes imperative to discuss their solution by numerical

methods.
as
In numerical methods, we donot proceed in the hope of finding a relation between

variables but we find the numerical values of the dependent variable for certain values of
independent variable.
g
.c
5.20 INITIAL-VALUE AND BOUNDARY-VALUE PROBLEMS (M.T.U. 2012)

w
Problems in which all the conditions are specified at the initial point only are called initial-
value problems. For example, the problem given by eqns. (1) and (2) is an initial value
w
problem.
Problems involving second and higher order differential equations, in which the condi-
w
tions at two or more points are specified, are called boundary-value problems.
To obtain a unique solution of nth order ordinary differential equation, it is necessary to
specify n values of the dependent variable and/or its derivative at specific values of independ-
ent variable.
5.21 SINGLE-STEP AND MULTI-STEP METHODS
The numerical solutions are obtained step-by-step through a series of equal intervals in the
independent variable so that as soon as the solution y has been obtained at x = xi , the next
step consists of evaluating yi+1 at x = xi+1. The methods which require only the numerical
www.cgaspirants.com
www.cgaspirants.com
value yi in order to compute the next value yi+1 for solving equation (1) given above are termed
as single-step methods.
The methods which require not only the numerical value yi but also at least one of the
past values yi–1, yi–2, ...... are termed as multi-step methods.
5.22 COMPARISON OF SINGLE-STEP AND MULTI-STEP METHODS
The single-step method has obvious advantages over the multi-step methods that use several
past values (yn, yn–1, ......, yn–p) and that require initial values (y1, y2, ......, yn) that have to be
calculated by another method.
The major disadvantage of single-step method is that they use many more evaluations
of the derivative to attain the same degree of accuracy compared with the multi-step methods.
m
5.23 NUMERICAL METHODS OF SOLUTION OF O.D.E.
o
Here, we will discuss various numerical methods of solving ordinary differential equations.
.c
We must know that these methods will yield the solution in one of the two forms:
(a) A series for y in terms of powers of x from which the value of y can be obtained by
direct substitution.
(b) A set of tabulated values of x and y.
n ts
Picard’s method belong to class (a) while those of Euler’s, Runge-Kutta, etc., belong to
ra
class (b). Methods which belong to class (b) are called step-by-step methods or marching
methods because the values of y are computed by short steps ahead for equal intervals of the
independent variable.
pi
In Euler’s and Runge-Kutta methods, the interval range h should be kept small hence
they can be applied for tabulating y only over a limited range.
as
5.24 PICARD’S METHOD OF SUCCESSIVE APPROXIMATIONS

g
Emile Picard (1856–1941) was a distinguished Professor of Mathematics at the University

.c
of Paris, France. He was famous for his researches on the Theory of Functions.
Consider the differential equation
w
dy
= f (x, y); y(x0) = y0 ...(1)
w
dx
Integrating eqn. (1) between the limits x0 and x and the corresponding limits y0 and y,
w
we get
z z
y
y0
dy =
x
x0
f ( x, y) dx
⇒ z
y – y0 =
x
x0
f ( x, y) dx
or, y = y0 + z x0
x
f ( x, y) dx ...(2)
In equation (2), the unknown function y appears under the integral sign. Such type of
equation is called integral equation.
www.cgaspirants.com
www.cgaspirants.com
This equation can be solved by the method of successive approximations or iterations.

To obtain the first approximation, we replace y by y0 in the RHS of eqn. (2).
Now, the first approximation is
y(1) = y0 + z x0
x
f ( x, y0 ) dx
The integrand is a function of x alone and can be integrated.

...(3)
For a second approximation, replace y0 by y(1) in f (x, y0) which gives
y(2) = y0 +
z
x0
x
f { x , y (1) } dx
Proceeding in this way, we obtain y(3), y(4), ......., y(n–1) and y(n) where
...(4)
m
y(n) = y0 + z x0
x
f { x , y ( n −1) } dx with y(x0) = y0
o
As a matter of fact, the process is stopped when the two values of y viz. y(n–1) and y(n) are
.c
same to the desired degree of accuracy.
Picard’s method is of theoretical value considerably. Practically, it is unsatisfactory
ts
because of the difficulties which arise in performing the necessary integrations. However,
each step gives a better approximation of the required solution than the preceding one.
n
ra
EXAMPLES
pi
Example 1. Use Picard’s method to obtain y for x = 0.2. Given:

dy
= x – y with initial condition y = 1 when x = 0.
as
dx
Sol. Here f(x, y) = x – y, x0 = 0, y0 = 1
We have first approximation,
g
z z
.c
x x x2
y(1) = y0 + f ( x, y0 ) dx = 1 + ( x − 1) dx = 1 – x +
0 0 2
w
Second approximation,
z z
w
x x
y(2) = y0 + f { x, y (1) } dx = 1 + { x − y (1) } dx
0 0
w
=1+
z FGH
0
x
x − 1+ x −
x2
2
I
JK
dx = 1 – x + x2 –
x3
6
Third approximation,
y(3) = y0 + z 0
x
f { x , y ( 2 ) } dx = 1 + z
0
x
{ x − y (2) } dx
=1+
z GHF
0
x
x − 1 + x − x2 +
x3
6
I
JK
dx = 1 – x + x2 –
x3 x4
3
+
24
www.cgaspirants.com
www.cgaspirants.com
Fourth approximation,
y(4) = y0 + z0
x
f { x, y (3) } dx = 1 +
z 0
x
{ x − y (3) } dx
=1+
z FGH
x
0
x − 1 + x − x2 +
x3 x4
3
−
24
dx
I
JK
x3 x4 x5
= 1 – x + x2 – + −
3 12 120
When x = 0.2, we get
y(1) = 0.82, y(2) = 0.83867, y(3) = 0.83740, y(4) = 0.83746
m
Since y(3) and y(4) are same upto four decimal places, therefore,
y(0.2) = 0.8374.
o
Example 2. Using Picard’s method of successive approximations, obtain a solution up
.c
dy
to fifth approximation of the equation = y + x such that y = 1 when x = 0.
dx
Sol. We have,
n
f(x, y) = y + x, x0 = 0, y0 = 1
ts
First approximation,
ra
y(1) = y0 + z x
f(x, y0) dx = y0 + z x
(x + y0) dx
pi
x0 0
=1+ zx
(x + 1) dx = 1 + x +
x2
as
0 2
zF x
z x
g
y(2) = y0 + f{x, y(1)} dx = 1 + {x + y(1)} dx

0 0
.c
=1+ z GH
x
x + 1+ x +
x2 I
JK dx = 1 + x + x 2 +
x3
w
0 2 6
w
y(3) = y0 + zF x
f{x, y(2)} dx = 1 + zI x
{x + y(2)} dx
w
0 0
=1+ z GH
x
0
x + 1 + x + x2 +
x3
6 JK dx = 1 + x + x 2 +
x3 x4
3
+
24
Fourth approximation,
y(4) = y0 + z x
0
f { x, y (3) } dx
=1+ z FGH
x
0
x + 1 + x + x2 +
x3 x4
3
+
24
dx
I
JK
x3 x4 x5
= 1 + x + x2 + + +
3 12 120
www.cgaspirants.com
www.cgaspirants.com
Fifth approximation,
y(5) = y0 +z x
0
f { x, y (4) } dx
=1+ z FGH
x
0
x + 1 + x + x2 +
x3 x4
3
+ +
x5
12 120
dx
I
JK
x3 x4 x5 x6
= 1 + x + x2 + + + +
3 12 60 720
dy y − x
Example 3. If = , find the value of y at x = 0.1 using Picard’s method. Given
dx y + x
m
that y(0) = 1.
Sol. First approximation,
o
y(1) = y0 +z x y0 − x
z FGH
x 1− xIJ
.c
dx = 1 + dx
0 y0 + x 0 1+ x K
=1+ z FGH
n x
0
2
1+ x
IJ
ts
− 1 dx = 1 – x + 2 log (1 + x)
K
ra
y(2) = 1 + x – 2 z0
x x dx
1 + 2 log (1 + x)
pi
which is difficult to integrate.

Thus, when x = 0.1, y(1) = 1 – 0.1 + 2 log (1.1) = 1.09062
as
Here in this example, only I approximation can be obtained and so it gives the approxi-
mate value of y for x = 0.1.
g
dy x2
.c
Example 4. Given the differential equation = 2 with the initial condition y = 0

dx y + 1
w
when x = 0. Use Picard’s method to obtain y for x = 0.25, 0.5 and 1.0 correct to three decimal
places.
w
Sol. The given initial value problem is

w
dy x2
= f(x, y) = 2 where y0 = 0 at x0 = 0
dx y +1
We have first approximation,
y(1) = y0 + z x
x0
f ( x, y0 ) dx = 0 + z0
x x2
0+1
1
dx = x3
3
...(1)
y(2) = y0 + z x
x0
f { x, y (1) } dx
www.cgaspirants.com
www.cgaspirants.com
=0+ z FG
0
x
I +1
x3
x2
2
dx = tan–1 x
3
3
H JK 3
– FG x IJ + ... = x
1 3 3
1 1 3 1 3 1 9
= x – x + ... ...(2)
3 3H3 K 3 81
x3
From (1) and (2), we see that y(1) and y(2) agree to the first term . To find the range of
3
1 3
values of x so that the series with the term x alone will give the result correct to three
m
3
decimal places, we put
o
1 9
x ≤ 0.0005
.c
81
which gives, x9 ≤ 0.0405 or x ≤ 0.7
Hence, y(0.25) =
n1
3
(0.25)3 = 0.005 ts
and y(0.5) =
1
3
(0.5)3 = 0.042
To find y(1.0), we make use of eqn. (2) which gives,

ra
1 1
y(1.0) = – = 0.321.
pi
3 81
5.25 PICARD’S METHOD FOR SIMULTANEOUS FIRST ORDER DIFFERENTIAL

as
EQUATIONS
dy dz
g
Let = φ(x, y, z) and = f(x, y, z)

dx dx
.c
be the simultaneous differential equations with initial conditions y(x0 ) = y0; z(x0) = z0.
w
Picard’s method gives
z z
w
x x
y(1) = y0 + φ( x, y0 , z0 ) dx ; z(1) = z0 + f ( x, y0 , z0 ) dx
x0 x0
w
y(2) = y0 + zx
x0
φ{ x, y (1) , z (1) } dx ; z(2) = z0 + z x
x0
f { x, y (1) , z (1) } dx
and so on as successive approximations.
5.26 PICARD’S METHOD FOR SECOND ORDER DIFFERENTIAL EQUATIONS
Consider the second order differential equation
d2 y FG dy IJ
dx 2 H
= f x, y,
dx K
www.cgaspirants.com
www.cgaspirants.com
dy
By putting = z, it can be reduced to two first order simultaneous differential equa-
dx
tions:
dy dz
= z and = f(x, y, z)
dx dx
which can be solved easily as explained in art. 5.25.
EXAMPLES
Example 1. Approximate y and z by using Picard’s method for the particular solution of
m
dy dz
= x + z, = x – y2 given that y = 2, z = 1 when x = 0. (G.B.T.U. 2011)
dx dx
o
Sol. Let φ(x, y, z) = x + z, f(x, y, z) = x – y2
.c
Here, x0 = 0, y0 = 2, z0 = 1
We have,
dy
dx
= φ(x, y, z)
n ⇒ y = y0 +
ts zx0
x
φ( x, y, z) dx ...(1)
Also,
dz
= f(x, y, z) ⇒ z = z0 + z x
f ( x, y, z) dx ...(2)
ra
dx x0
z z
pi
x x
y(1) = y0 + φ( x, y0 , z0 ) dx = 2 + ( x + z0 ) dx
x0 0
as
=2+ z x
0
( x + 1) dx = 2 + x +
x2
2
g
z x
f ( x, y0 , z0 ) dx = 1 +
z x
( x − y0 2 ) dx
.c
and z(1) = z0 +
x0 0
z
w
x x2
=1+ ( x − 4) dx = 1 – 4x +
0 2
w
z x
φ{ x, y (1) , z (1) } dx = 2 +
z x
w
y(2) = y0 + { x + z (1) } dx
x0 0
=2+
z FGH
x
0
x + 1 − 4x +
x2
2
I
JK
3
dx = 2 + x – x2 +
2
x3
6
z(2) = z0 +
z x
x0
f { x , y (1) , z (1) } dx
LM OP
z MN F x I
2
x 2
x − G2 + x +
2 JK
=1+
0 H PQ dx
3 2 x4 x5
= 1 – 4x – x – x3 – – .
2 4 20
www.cgaspirants.com
www.cgaspirants.com
dy dz
Example 2. Solve by Picard’s method, the differential equations = z, = x3 (y + z)
dx dx
1
where y = 1, z= at x = 0. Obtain the values of y and z from III approximation when
2
x = 0.2 and x = 0.5.
Sol. Let φ(x, y, z) = z,f(x, y, z) = x3(y + z)
1
Here x0 = 0, y0 = 1, z0 =
2
y(1) = y0 + z x
φ( x, y0 , z0 ) dx = 1 +
z x
z0 dx = 1 +
1
x
m
0 0 2
z(1) = z0 + z x
f ( x, y0 , z0 ) dx =
1
+ zx
x 3 ( y0 + z0 ) dx =
1 3 4
+ x .
o
0 2 0 2 8
.c
y(2) = 1 + z0
x
z (1) dx = 1 + z FGH
ts
0
x 1 3 4 x
+ x dx = 1 + +
2 8 2 40
IJ
3 5
x
K
1
z x
n
z(2) = + x 3 { y (1) + z (1) } dx
2 0
ra
=
1
2
+ z0
x
x3
FG 3 + x + 3 x IJ dx = 1 + 3 x
H2 2 8 K 2 8
4 4
+
x5
+
10 64
3 8
x
pi
z x
z x F 1 + 3x 4
x5 3x8 I
as
y(3) = 1 +
0
z (2) dx = 1 +
0
GH 2 8 +
10
+
64
dx JK
6
x9
g
x 3 5 x
=1+ + x + +
2 40 60 192
.c
z(3) =
1
+ z x
x 3 { y (2) + z (2) } dx
w
2 0
1
zx RS 3 + x + 3 x 7 5 3 8 UV
w
= + x3 4
+ x + x dx
2 0 T2 2 8 40 64 W
w
1 3 4 x5 3 8 7 9 1 12
= + x + + x + x + x
2 8 10 64 360 256
When x = 0.2
3 (0.2) 6 1
y(3) = 1 + 0.1 + (0.2)5 + + (0.2)9
40 60 192
= 1.100024 (leaving higher terms)
1 3 (0.2) 5 3 7 1
z(3) = + (0.2)4 + + (0.2)8 + (0.2)9 + (0.2)12
2 8 10 64 360 256
= 0.500632 (leaving higher terms)
www.cgaspirants.com
www.cgaspirants.com
When x = 0.5
0.5 3 (0.5) 6 1
y(3) = 1 + + (0.5)5 + + (0.5)9 = 1.25234375
2 40 60 192
1 3 (0.5) 5 3 7 1
z(3) = + (0.5)4 + + (0.5)8 + (0.5)9 + (0.5)12
2 8 10 64 360 256
= 0.5234375.
Example 3. Use Picard’s method to approximate y when x = 0.1 given that
d2 y dy dy
2 + 2x + y = 0 and y = 0.5, = 0.1 when x = 0.
dx dx dx
m
d2 y dy
Sol. We have, + 2x +y=0 ...(1)
o
2
dx dx
.c
dy dz
Let = z so that eqn. (1) becomes + 2xz + y = 0
dx dx
Now the equations to be solved are
dy
dx
n
= f(x, y, z) = z
ts ...(2)
ra
dz
= φ(x, y, z) = – (2xz + y) ...(3)
dx
pi
with the conditions y0 = 0.5, z0 = 0.1 at x0 = 0

First approximation
z z
as
x x
y(1) = y0 + f(x, y0, z0) dx = 0.5 + z0 dx = 0.5 + 0.1 x
x0 0
z z
g
x x
z(1) = z0 + φ(x, y0, z0) dx = 0.1 + [– (2x z0 + y0)] dx
.c
x0 0
z x
w
= 0.1 – (0.2x + 0.5) dx = 0.1 – 0.5x – 0.1x2

0
w
y(2) = y0 + z x
f{x, y(1), z(1)} dx = 0.5 + z x
z(1) dx
w
0 0
= 0.5 + z 0
x
(0.1 – 0.5x – 0.1x2) dx
0.1 3
= 0.5 + 0.1x – 0.25x2 – x
3
z(2) = z0 + z0
x
φ {x, y(1), z(1)} dx = 0.1 – z
0
x
[2x z(1) + y(1)] dx
= 0.1 – z 0
x
[2x(0.1 – 0.5x – 0.1x2) + (0.5 + 0.1x)] dx
www.cgaspirants.com
www.cgaspirants.com
= 0.1 – z0
x
(0.3x – x2 – 0.2x3 + 0.5) dx
x 2 x 3 0.1 x 4
= 0.1 – 0.5x – 0.3 + +
2 3 2
Third approximation
y(3) = y0 + zF
0
x
f{x, y(2), z(2)} dx = 0.5 + z0
x
z(2) dx
= 0.5 + z GH0
x
0.1 − 0.5 x − 0.15 x 2 +
x3
3
+ 0.05 x 4
I
JK dx
x4
m
= 0.5 + 0.1x – 0.25x2 – 0.05x3 + + 0.01x5
12
z x
z x
o
z(3) = z0 + φ {x, y(2), z(2)} dx = 0.1 – [2x z(2) + y(2)] dx
0 0
.c
= 0.1 – z LMMN
x F
GH
2 x 0.1 − 0.5 x − 0.15 x 2 +
x3 I FG
JK H
+ 0.05 x 4 + 0.5 + 0.1x − 0.25 x 2 −
0.1 3
x
IJ OP dx
K PQ
0
= 0.1 – 0.5x + 0.15x2 –

2.5 3
6
x + 0.2x4 +
2 5 0.1 6
15
n
x +
6
x
3
ts 3
ra
Now when x = 0.1, y(1) = 0.51, y(2) = 0.50746667, y(3) = 0.50745933
Since y(2) and y(3) are same up to four decimal places hence y(0.1) = 0.5074.
pi
ASSIGNMENT
as
dy
1. Find the solution of = 1 + xy, y (0) = 1 in the interval (0, 0.5) correct to 3 decimal places taking
g
dx
h = 0.1.
.c
dy
2. Solve numerically = 2x – y, y(0) = 0.9 at x = 0.4 by Picard’s method with three iterations and
w
dx
compare the result with the exact value.
w
dy
3. Apply Picard’s method to find the solution of the initial value problem = y – x, y (0) = 2. Show
dx
w
that the iterative solution approaches the exact solution. (M.T.U. 2012)
4. Use Picard’s method to approximate the value of y when x = 0.1 given that y = 1 when x = 0 and
dy
= 3x + y2. (M.T.U. 2013)
dx
dy
5. (i) For the differential equation = x – y2, y(0) = 0, calculate y(0.2) by Picard’s method to
dx
third approximations and round off the value at 4th place of decimals.
dy
(ii) Find an approximate value of y when x = 0.1 if = x – y2 and y = 1 at x = 0 using Picard’s
dx
method.
www.cgaspirants.com
www.cgaspirants.com
dy
6. Employ Picard’s method to find y (0.2) and y (0.4) given that = 1 + y2 and y (0) = 0.
dx
dy
7. Employ Picard’s method to obtain the solution of = x2 + y2 for x = 0.4 correct to four places
dx
of decimal given that y = 0 when x = 0.
dy
8. Find y(0.2) if = log10 (x + y); y(0) = 1. Use Picard’s method.
dx
2
FG IJ
9. Solve by Picard’s method: d y = x3 dy + y where y = 1, dy = 1 when x = 0, obtain the results
dx2 dx H dx 2K
up to third approximation.
m
Answers
o
1. x: 0 0.1 0.2 0.3 0.4 0.5
.c
y: 1 1.105 1.222 1.355 1.505 1.677
2. 0.7432, 0.7439
3. y = 2 + 2x +
x2
2
+
x3 x4
6
−
24
5 2 4 3 3 4 9 5
n ts
(up to third approx.); exact solution: y = 1 + x + ex
4. y=1+x+ x + x + x + x (up to second approx.)

ra
2 3 4 20
5. (i) 0.019984, 0.0200 (ii) 0.9138 6. 0.2027, 0.4227

pi
7. 0.0213 8. 1.0082
as
x (2) x 3 x5 (3) x 3 x5 x6 x9
9. y(1) = 1 + ,y =1+ + ,y =1+ + + + .
2 2 40 2 40 60 192
g
5.27 EULER’S METHOD [G.B.T.U. (MCA) 2010]

.c
It is the simplest one-step method and has a limited application because of its low accuracy.
w
This method yields solution of an ordinary diff. eqn. in the form of a set of tabulated values.
In this method, we determine the change Δy is y
w
)
corresponding to small increase in the argument x. Consider Y ,y1
)
(f x 1
g(x
the differential equation

w
pe Q2
y=
slo
dy y 0)
= f ( x, y) , y(x0) = y0 ...(1) f(x 0
,
dx e
slop
y2
Let y = g(x) be the solution of (1). Let x0, x1, x2 ...... be Q1
equidistant values of x. P0
y1
In this method, we use the property that in a small y0
interval, a curve is nearly a straight line. Thus at the point X
(x0, y0), we approximate the curve by the tangent at the point O x0 x1 x2
(x0, y0).
www.cgaspirants.com
www.cgaspirants.com
The equation of the tangent at P0(x0, y0) is
FG dy IJ
y – y0 = H dx K P0
( x – x0 ) = f(x , y ) (x – x )
0 0 0
⇒ y = y0 + (x – x0) f(x0, y0) ...(2)

This gives the y-coordinate of any point on the tangent. Since the curve is approximated
by the tangent in the interval (x0, x1), the value of y on the curve corresponding to x = x1 is
given by the above value of y in eqn. (2) approximately.
Putting x = x1(= x0 + h) in eqn. (2), we get
y1 = y0 + hf(x0, y0)
m
Thus Q1 is (x1, y1)
Similarly, approximating the curve in the next interval (x1, x2) by a line through
o
Q1(x1, y1) with slope f(x1, y1), we get
y2 = y1 + hf(x1, y1)
.c
In general, it can be shown that,
yn+1 = yn + hf(xn, yn)
This is called Euler’s formula.

n ts
ra
dy
A great disadvantage of this method lies in the fact that if changes rapidly over an
dx
interval, its value at the beginning of the interval may give a poor approximation as compared
pi
to its average value over the interval and thus the value of y calculated from Euler’s method
may be in much error from its true value. These errors accumulate in the succeeding intervals
as
and the value of y becomes much erroroneous ultimately.

Note. In Euler’s method, the curve of actual solution y = g(x) is approximated by a sequence of short
lines. The process is very slow. If h is not properly chosen, the curve P0Q1Q2 ...... of short lines representing
g
numerical solution deviates significantly from the curve of actual solution.

.c
5.28 IMPROVED EULER’S METHOD

w
The improved Euler’s method gives greater improvement in accuracy over the original Euler’s
w
method. Here the core idea is that we use a line through (x0, y0) whose slope is the average of
the slopes at (x0, y0) and (x1, y1(1)) where y1(1) = y0 + hf(x0, y0). This line approximates the curve
w
in the interval (x0, x1).

Geometrically, if L1 is the tangent at (x0, y0), L2 is Y L
(x1, y1)
a line through (x1, y1(1)) of slope f(x1, y1(1)) and L is the B
line through (x1, y1(1)) but with a slope equal to the aver- L2
age of f(x0, y0) and f(x1, y1(1)) then the line L through L
(x0, y0) and parallel to L is used to approximate the curve (x0, y0)
in the interval (x0, x1). Thus the ordinate of the point B A (1)
L1
(x1, y1 )
will give the value of y1. Now, the eqn. of the line AL is
given by
y1 = y0 + (x1 – x0)
LM f ( x , y ) + f ( x , y
0 0 1 1
(1)
) OP X
MN PQ
O x0 x1
2
www.cgaspirants.com
www.cgaspirants.com
= y0 + h
LM f (x , y ) + f (x , y ) OP
0 0 1 1
(1)
...(1)
MN 2 PQ
A generalized form of improved Euler’s formula is
h
y1(n+1) = y0 + [f(x0, y0) + f{x1, y1(n)}] ; n = 0, 1, 2, ...... ...(2)
2
where y1(n) is the nth approximation to y1.

The above iteration formula can be started by choosing y1(1) from Euler’s formula
y1(1) = y0 + hf(x0, y0) ...(3)
Since this formula attempts to correct the values of yn+1 using the predicted value of
m
yn+1 (by Euler’s method), it is classified as a one-step predictor-corrector method.
o
5.29 MODIFIED EULER’S METHOD
.c
In this method, the curve in the interval (x0, x1), where
(x1, y1)
x1 = x0 + h is approximated by the line through (x0, y0)
ts
Y L
B
with slope f
RSx h
+
h UV
, y0 + f ( x0 , y0 ) , which is the
T W
n
0
2 2 L
slope at the middle point whose abscissa is the aver-

ra
age of x0 and x1. (x0, y0)

P
In the adjoining figure, line AL through A(x0, y0) A
L1
pi
which is parallel to the line PL with slope y0 y0 + h f(x0, y0) y1

2
RS h h UV
as
f x0 + , y0 + f ( x0 , y0 ) approximates the curve in

T 2 2 W O x0 x0 + h
2
x1
X
the interval (x0, x1). The ordinate at x = x1, meeting the

g
line L at B, will give the value of y1.

.c
The equation for line AL is

LM RS h h UVOP
w
y – y0 = (x – x0) f x0 + , y0 + f ( x0 , y0 )
N T 2 2 WQ ...(1)
w
Putting x = x1 in (1), we get

LM RS h h UVOP
w
y1 = y0 + (x1 – x0) f x0 + , y0 + f ( x0 , y0 )
N T 2 2 WQ
y1 = y0 + h f x0 +
RS h h
, y0 + f ( x0 , y0 )
UV
⇒
T 2 2 W
Proceeding in the same way, we obtain
yn+1 = yn + h f xn +
RS h h
, yn + f ( xn , yn )
UV ...(2)
T 2 2 W
which is the generalized form of modified Euler’s formula.
www.cgaspirants.com
www.cgaspirants.com
EXAMPLES
dy y – x
Example 1. Given = with y = 1 for x = 0. Find y approximately for x = 0.1 by
dx y + x
Euler’s method.
Sol. We have
dy y– x
= f(x, y) = ; x0 = 0, y0 = 1, h = 0.1
dx y+ x
Hence the approximate value of y at x = 0.1 is given by
m
y1 = y0 + hf(x0, y0) | Using yn+1 = yn + hf (xn, yn)
FG 1 – 0 IJ = 1.1
o
= 1 + (0.1)
H 1 + 0K
.c
Much better accuracy is obtained by breaking up the interval 0 to 0.1 into five steps.
The approximate value of y at xA = 0.02 is given by,
y1 = y0 + hf(x0, y0) = 1 + (0.02)

n ts
FG 1 – 0 IJ = 1.02
H 1 + 0K
ra
At xB = 0.04, y = y + hf(x , y ) = 1.02 + (0.02) G

F 1.02 – 0.02 IJ = 1.0392
2 1 1 1
H 1.02 + 0.02 K
pi
At xC = 0.06, y = 1.0392 + (0.02) G

F 1.0392 – 0.04 IJ = 1.0577
3
H 1.0392 + 0.04 K
as
At xD = 0.08, y = 1.0577 + (0.02) G

F 1.0577 – 0.06 IJ = 1.0756
4
H 1.0577 + 0.06 K
g
y = 1.0756 + (0.02) G
F 1.0756 – 0.08 IJ = 1.0928
.c
At xE = 0.1, 5
H 1.0756 + 0.08 K
w
Hence y = 1.0928 when x = 0.1

w
E
Y
w
C
B
A
X
O A¢ B¢ C¢ D¢ E¢
www.cgaspirants.com
www.cgaspirants.com
dy
Example 2. Solve the equation = 1 – y with the initial condition x = 0, y = 0 using
dx
Euler’s algorithm and tabulate the solutions at x = 0.1, 0.2, 0.3.
Sol. Here, f(x, y) = 1 – y
Taking h = 0.1, x0 = 0, y0 = 0, we obtain
y1 = y0 + hf(x0, y0) = 0 + (0.1) (1 – 0) = 0.1
∴ y(0.1) = 0.1
Again, y2 = y1 + hf(x1, y1) = 0.1 + (0.1) (1 – 0.1) = 0.1 + 0.09 = 0.19
∴ y(0.2) = 0.19
Again, y3 = y2 + hf(x2, y2) = 0.19 + (0.1) (1 – 0.19) = 0.271
m
∴ y(0.3) = 0.271
Tabulated values are
o
.c
x y(x)
0 0
0.1
0.2
n
0.3
ts 0.1
0.19
0.271
ra
Example 3. Use Euler’s method to obtain an approximate value of y(0.4) for the equation
y′ = x + y, y(0) = 1 with h = 0.1.
pi
Sol. Here, f(x, y) = x + y

y1 = y0 + h f(x0, y0) = 1 + (0.1) (0 + 1) = 1.1
as
∴ y(0.1) = 1.1
Again, y2 = y1 + h f(x1, y1) = 1.1 + (0.1) (0.1 + 1.1) = 1.22
g
∴ y(0.2) = 1.22
.c
Again, y3 = y2 + h f(x2, y2) = 1.22 + (0.1) (0.2 + 1.22) = 1.362

∴ y(0.3) = 1.362
w
Again, y4 = y3 + h f(x3, y3) = 1.362 + (0.1) (0.3 + 1.362) = 1.5282

∴ y(0.4) = 1.5282
w
Example 4. Solve the following differential equation using Euler’s method from x = 0 to
x = 0.2 when h = 0.05.
w
dy
+ xy = 0, y(0) = 1
dx
Sol. Here, f(x, y) = – xy
y1 = y0 + h f (x0, y0) = 1 + (0.05) {– (0 × 1)} = 1
∴ y (0.05) = 1
Again, y2 = y1 + h f (x1, y1) = 1 + (0.05) {– (0.05 × 1)} = 0.9975
∴ y(0.1) = 0.9975
Again, y3 = y2 + h f (x2, y2) = 0.9975 + (0.05) {– (0.1 × 0.9975)} = 0.9925125
∴ y(0.15) = 0.9925125
www.cgaspirants.com
www.cgaspirants.com
Again, y4 = y3 + h f(x3, y3) = 0.9925125 + (0.05) {– (0.15) (0.9925125)}

= 0.985068656
∴ y(0.2) = 0.985068656
Hence, in tabular form, we can write
x 0 0.05 0.10 0.15 0.2
y 1 1 0.9975 0.9925125 0.985068656
Example 5. Find y(0.1) using improved Euler’s method and then y(0.2) by using modified
dy
Euler’s method, given that = log (x + y), y(0) = 1.0 (U.P.T.U. 2007)
dx
m
Sol. Here f(x, y) = log (x + y)
Also, x0 = 0, y0 = 1, h = 0.1
o
Using improved Euler’s method, we get
.c
y1(1) = y0 + h f(x0, y0) = 1 + (0.1) log (0 + 1) = 1
y1(2) = y0 +
n h
2 ts
[f(x0, y0) + f{x1, y1(1)}]
(0.1)
=1+ [0 + log(0.1 + 1)] = 1.0047655
ra
2
h
y1(3) = y0 + [f(x0, y0) + f{x1, y1(2)}]
pi
2
(0.1)
as
=1+ [0 + log (0.1 + 1.0047655)] = 1.0049816

2
h
g
y1(4) = y0 + [f(x0, y0) + f{x1, y1(3)}]

2
.c
(0.1)
=1+ [0 + log(0.1 + 1.0049816)] = 1.0049914
w
2
Since y1(3) and y1(4) are same up to four decimal places
w
∴ y1 = y(0.1) = 1.0049
Again, x1 = 0.1, y1 = 1.0049, h = 0.1
w
Using modified Euler’s method, we get
y2 = y1 + h f x1 +
RS h h
, y1 + f ( x1 , y1 )
UV
T 2 2 W
= 1.0049 + (0.1) log x1 +
RS h h
+ y1 + f ( x1 , y1 )
UV
T 2 2 W
= 1.0049 + (0.1) log [0.1 + 0.05 + 1.0049 + (0.05) log (0.1 + 1.0049)]
= 1.0197323
∴ y(0.2) = 1.0197323
www.cgaspirants.com
www.cgaspirants.com
Example 6. Apply modified Euler’s method to solve

dy
= ex + xy, y(0) = 0 to compute y(0.1) and y(0.2).
dx
Sol. Here, f(x, y) = ex + xy
We have, x0 = 0, y0 = 0, h = 0.1
Using modified Euler’s method, we obtain
RS
y1 = y0 + h f x0 +
h h
, y0 + f ( x0 , y0 )
UV
T 2 2 W
= 0 + (0.1) [e0.05 + (0.05) (0.05) (e0 + 0)] = 0.105377
∴ y(0.1) = 0.105377
m
Again,
RS
y2 = y1 + h f x1 +
h h
, y1 + f ( x1 , y1 )
UV
T W
o
2 2
= 0.105377 + (0.1) [e0.15 + (0.15) {(0.105377) + (0.05) (e0.1 + 0.0105377)}]
.c
= 0.105377 + (0.1) [e0.15 + (0.15) (0.16116243)] = 0.22397786
∴ y(0.2) = 0.22397786. n
ASSIGNMENT
ts
ra
1. Using Euler’s method, compute y(0.04) for the differential equation

dy
pi
= – y; y(0) = 1. Take h = 0.01. (U.P.T.U. 2007)

dx
dy
as
2. Apply Euler’s method to solve = x + y; y(0) = 0 choosing h = 0.2 and compute y(0.4), y(0.6),
dx
y(0.8) and y(1.0).
g
dy
3. If = 1 + y2, y(0) = 1, find y(0.4) by using Euler’s method. Take h = 0.2.
.c
dx
[G.B.T.U. (C.O.) 2011]
w
dy
4. Given that = log10 (x + y); y(0) = 1
dx
w
Find y(0.2) and y(0.5) using improved Euler’s method.

w
dy
5. Using improved Euler’s method, obtain a solution of the equation =x+| y | = f(x, y)
dx
with initial condition y = 1 at x = 0 for the range 0 ≤ x ≤ 0.6 in steps of 0.2.
6. Use Euler’s modified method to obtain y(0.25) given that y′ = 2xy, y(0) = 1
7. Solve y′ = x2 + y, y(0) = 1 to obtain y(0.02) and y(0.04) using Euler’s modified method.
dy 2x
8. Solve =y– ; y(0) = 1 in the range 0 ≤ x ≤ 0.2 using
dx y
(i) Improved Euler’s method (ii) Modified Euler’s method.
www.cgaspirants.com
www.cgaspirants.com
Answers
1. y(0.04) = 0.960596
2. y(0.4) = 0.04, y(0.6) = 0.128, y(0.8) = 0.2736, y(1.0) = 0.48832
3. y(0.4) = 1.992 4. y(0.2) = 1.0082, y(0.5) = 1.0490
5. y(0.2) = 1.2309, y(0.4) = 1.5253, y(0.6) = 1.8861
6. y(0.25) = 1.0625 7. y(0.02) = 1.0202, y(0.04) = 1.0408
8.
x Improved Euler’s Modified Euler’s

method method
0 1 1
m
0.1 1.095 1.0954762
0.2 1.1828 1.1832984
o
.c
5.30 RUNGE-KUTTA METHODS
ts
More efficient methods in terms of accuracy were developed by two German Mathematicians
Carl Runge (1856–1927) and Wilhelm Kutta (1867–1944). These methods are well-known
n
as Runge-Kutta methods. They are distinguished by their orders in the sense that they
agree with Taylor’s series solution upto terms of hr, where r is the order of the method.
ra
These methods donot demand prior computation of higher derivatives of y(x) as in Taylor’s
method. In place of these derivatives, extra values of the given function f(x, y) are used.
pi
Fourth order Runge-Kutta method is widely used for finding the numerical solutions of
linear or non-linear ordinary differential equations.
as
Runge-Kutta methods are referred to a single step methods. The major disadvantage of
Runge-Kutta methods is that they use many more evaluations of the derivative f(x, y) to obtain
the same accuracy compared with multi-step methods. A class of methods known as Runge-
g
Kutta methods combines the advantage of high order accuracy with the property of being one
.c
step.
5.30.1. First Order Runge-Kutta Method
w

w
dy
= f(x, y); y (x0) = y0 ...(1)
dx
w
Euler’s method gives

y1 = y0 + hf(x0, y0) = y0 + hy0′ ...(2)
Expanding by Taylor’s series, we get
h2
y1 = y(x0 + h) = y0 + hy0′ + y ″ + ... ...(3)
2! 0
Comparing (2) and (3), it follows that Euler’s method agrees with Taylor’s series solution
up to the term in h. Hence Euler’s method is the first order Runge-Kutta method.
www.cgaspirants.com
www.cgaspirants.com
5.30.2. Second Order Runge-Kutta Method

y′ = f(x, y) with the initial condition y(x0) = y0
Let h be the interval between equidistant values of x then in second order Runge-Kutta
method, the first increment in y is computed from the formulae
k1 = hf (x0, y0)
UV
k2 = hf(x0 + h, y0 + k1)
Δy = 1
2
(k1 + k2)
W
taken in the given order.
m
Then, x1 = x0 + h
o
1
y1 = y0 + Δy = y0 + 2
(k1 + k2)
.c
In a similar manner, the increment in y for the second interval is computed by means
of the formulae,
k1 = hf (x1, y1)
n
k2 = hf (x1 + h, y1 + k1)
ts
ra
1
Δy = 2
(k1 + k2)
and similarly for the next intervals.
pi
The inherent error in the second order Runge-Kutta method is of order h3.
5.30.3. Third Order Runge-Kutta Method

as
This method gives the approximate solution of the initial value problem
dy
g
= f (x, y); y(x0) = y0 as

dx
.c
y1 = y0 + Δy ...(1)
w
h
where Δy = (k + 4k2 + k3)
6 1
w
Here, k1 = f (x0, y0)

w
RS
k2 = f x0 +
h k
, y0 + 1
UV
T 2 2 W
k3 = f (x0 + h, y0 + k′); k′ = hf (x0 + h, y0 + k1)
Formula (1) can be generalised for successive approximations. Expression in (1) agrees
with Taylor’s series expansion for y1 up to and including terms in h3. This method is also
known as Runge’s method.
5.31 FOURTH ORDER RUNGE-KUTTA METHOD (U.P.T.U. 2015)
It is one of the most widely used methods and is particularly suitable in cases when the
computation of higher derivatives is complicated.
www.cgaspirants.com
www.cgaspirants.com
Consider the differential equation y′ = f(x, y) with the initial condition y(x0) = y0. Let h
be the interval between equidistant values of x then the first increment in y is computed from
the formulae
k1 = hf (x0, y0)
FG h k IJ
U|
k2 = hf x0 +
H 2
, y0 + 1
2 K |V
F
= hf G x
h 2 k I
+ J ||
k3
H0
2
+, y0
2K
....(1)
k4 = hf (x0 + h, y0 + k3 ) W
1
Δy = (k1 + 2k2 + 2k3 + k4 )
6
m
taken in the given order.
Then, x1 = x0 + h and y1 = y0 + Δy
o
In a similar manner, the increment in y for the second interval is computed by means of
.c
the formulae
k1 = hf (x1, y1)
FG
H
k2 = hf x1 +
F
n
h
h
2
k
, y1 + 1
2
IJ
k I
K ts
= hf G x 2
+ J
k3
H +, y1
2K
ra
1
2
k4 = hf (x1 + h, y1 + k3)
1
pi
Δy = (k1 + 2k2 + 2k3 + k4)

6
and similarly for the next intervals.
as
This method is also termed as Runge-Kutta’s method simply.

It is to be noted that the calculations for the first increment are exactly the same as for
g
any other increment. The change in the formula for the different intervals is only in the values
of x and y to be substituted. Hence to obtain Δy for the nth interval, we substitute xn–1, yn–1, in
.c
the expressions for k1, k2, etc.

w
The inherent error in the fourth order Runge-Kutta method is of order h5.
w
5.32 RUNGE-KUTTA METHOD FOR SIMULTANEOUS FIRST ORDER EQUATIONS
Consider the simultaneous equations

w
dy
= f1(x, y, z) ...(1)
dx
dz
= f2 (x, y, z) ...(2)
dx
with the initial conditions y(x0) = y0 and z(x0) = z0. Now, starting from (x0, y0, z0), the increments
k and l in y and z are given by the following formulae:
k1 = hf1(x0, y0, z0); l1 = hf2(x0, y0, z0)
FG h k l
, y0 + 1 , z0 + 1 ;
IJ FG h k l
, y0 + 1 , z0 + 1
IJ
k2 = hf1 x0 +
H 2 2 2 K H
l2 = hf2 x0 +
2 2 2 K
www.cgaspirants.com
www.cgaspirants.com
FG
k3 = hf1 x0 +
h k l
, y0 + 2 , z0 + 2 ;
IJ FG
l3 = hf2 x0 +
h k l
, y0 + 2 , z0 + 2
IJ
H 2 2 2 K H 2 2 2 K
k4 = hf1(x0 + h, y0 + k3 , z0 + l3); l4 = hf2(x0 + h, y0 + k3 , z0 + l3)
1 1
k= (k + 2k2 + 2k3 + k4); l= (l + 2l2 + 2l3 + l4)
6 1 6 1
Hence, y1 = y0 + k, z1 = z0 + l
To compute y2, z2, we simply replace x0, y0, z0 by x1, y1, z1 in the above formulae.
5.33 RUNGE-KUTTA METHOD FOR SECOND ORDER DIFFERENTIAL EQUATIONS
m
Consider a second order differential equation
d2 y dy FG IJ
o
dx 2 = φ x, y,
dx H K ...(1)
.c
with initial conditions y(x0) = y0, y′(x0) = y0′ ...(2)
d 2 y dz
Let
dy
dx
= z so that =
dx 2 dx
.
Substituting in equation (1), we get

n ts
ra
dz
= φ (x, y, z)
dx
with initial conditions y(x0) = y0, z(x0) = z0
pi
Hence, the problem is reduced to solving the simultaneous equations

as
dy dz
= z = f1 (x, y, z) and = f2 (x, y, z)
dx dx
subject to y(x0) = y0, z(x0) = z0.
g
Above simultaneous equations can be solved as explained in art. 5.32.

.c
w
EXAMPLES
w
dy
Example 1. Solve the equation = x + y with initial condition y(0) = 1 by Runge-Kutta
dx
w
rule, from x = 0 to x = 0.4 with h = 0.1. (M.T.U. 2014)

Sol. Here f(x, y) = x + y, h = 0.1, x0 = 0, y0 = 1
We have,
k1 = hf (x0, y0) = 0.1 (0 + 1) = 0.1
FG
k2 = hf x 0 +
h k
, y0 + 1
IJ = 0.1 (0.05 + 1.05) = 0.11
H 2 2 K
F
= hf G x
h
, y0
k I
+ J = 0.1105
2
k3
H 0 +
2 2K
k4 = hf (x0 + h, y0 + k3) = 0.12105
www.cgaspirants.com
www.cgaspirants.com
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.11034
6 1
Thus, x1 = x0 + h = 0.1 and y1 = y0 + Δy = 1.11034
Now for the second interval, we have
k1 = hf (x1, y1) = 0.1 (0.1 + 1.11034) = 0.121034
FG
k2 = hf x1 +
h k
, y1 + 1
IJ = 0.13208
H 2 2 K
k3
F
= hf G x
h
, y1
k I
+ J = 0.13263
2
H 1 +
2 2K
m
k4 = hf (x1 + h, y1 + k3) = 0.14429
1
o
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.132460
6 1
.c
Hence x2 = 0.2 and y2 = y1 + Δy = 1.11034 + 0.13246 = 1.24280
Similarly, for finding y3, we have
∴ y3 = 1.3997
n ts
k1 = hf (x2, y2) = 0.14428, k2 = 0.15649, k3 = 0.15710, k4 = 0.16999
and for y4 = y(0.4), we calculate

ra
k1 = 0.16997, k2 = 0.18347, k3 = 0.18414, k4 = 0.19838

∴ y4 = 1.5836.
pi
dy
Example 2. Given = y – x, y(0) = 2. Find y(0.1) and y(0.2) correct to four decimal
dx
as
places. (Use both second and fourth order methods) (U.P.T.U. 2014, 2015)
Sol. By second order Method
g
To find y(0.1)
.c
Here y′ = f (x, y) = y – x, x0 = 0, y0 = 2 and h = 0.1

w
Now, k1 = hf (x0, y0) = 0.1(2 – 0) = 0.2

k2 = hf (x0 + h, y0 + k1) = 0.21
w
1
∴ Δy = (k1 + k2) = 0.205
2
w
Thus, x1 = x0 + h = 0.1 and y1 = y0 + Δy = 2.205

To find y(0.2) we note that,
x1 = 0.1, y1 = 2.205, h = 0.1
For second interval, we have,
k1 = hf (x1, y1) = 0.2105
k2 = hf (x1 + h, y1 + k1) = 0.22155
1
∴ Δy = (k1 + k2) = 0.216025
2
Thus, x2 = x1 + h = 0.2 and y2 = y1 + Δy = 2.4210
Hence y(0.1) = 2.205, y(0.2) = 2.421.
www.cgaspirants.com
www.cgaspirants.com
By fourth order method

y′ = f(x, y) = y – x, x0 = 0, y0 = 2 and h = 0.1
k1 = hf(x0, y0) = 0.1 (2 – 0) = 0.2
FG h k
, y0 + 1
IJ = 0.205
H
k2 = hf x0 +
2 2 K
F
= hf G x
h k I
+ J = 0.20525
2
k3
H 0 +
2
, y0
2K
and k4 = hf (x0 + h, y0 + k3) = 0.210525
1
m
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.2052
6 1
Thus, x1 = x0 + h = 0 + 0.1 = 0.1
o
y1 = y0 + Δy = 2 + 0.2052 = 2.2052
.c
Now to determine y2 = y(0.2), we note that
For II interval,
x1 = x0 + h = 0.1, y1 = 2.2052, h = 0.1
k1 = hf (x1, y1) = 0.21052
n ts
FG h k IJ = 0.21605
ra
, y1 + 1
H
k2 = hf x1 +
2 2 K
F
= hf G x
k I
+ J = 0.216323
pi
2
k3
H 1 + h / 2, y1
2K
and k4 = hf (x1 + h, y1 + k3) = 0.221523
as
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.21613
6 1
g
Thus, x2 = x1 + h = 0.1 + 0.1 = 0.2

.c
and y2 = y1 + Δy = 2.2052 + 0.21613 = 2.4213

Hence, y(0.1) = 2.2052, y(0.2) = 2.4213.
w
Example 3. Given the initial value problem: y′ = 1 + y2, y(0) = 0

w
Find y(0.6) by Runge-Kutta fourth order method taking h = 0.2. (U.P.T.U. 2008, 2006)
Sol. Here, f(x, y) = 1 + y2, h = 0.2, x0 = 0, y0 = 0
w
We have, k1 = h f (x0, y0) = 0.2

FG h k
, y0 + 1
IJ = 0.202
H
k2 = h f x0 +
2 2 K
F
= h f Gx
h k I
+ J = 0.2020402
2
k3
H0 +
2
, y0
2K
k4 = h f (x0 + h, y0 + k3) = 0.2081640
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.2027074
6 1
www.cgaspirants.com
www.cgaspirants.com
Thus, y1 = y(0.2) = y0 + Δy = 0.2027074, x1 = x0 + h = 0.2

Again, f(x, y) = 1 + y2, h = 0.2, x1 = 0.2, y1 = 0.2027074
We have, k1 = h f (x1, y1) = 0.208218
FG h k
, y1 + 1
IJ = 0.218827
H
k 2 = h f x1 +
2 2 K
F
= h f Gx
h k I
+ J = 0.219484
2
k3
H 1 +
2
, y1
2K
k4 = h f (x1 + h, y1 + k3) = 0.235649
1
m
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.22008
6 1
Thus, y2 = y(0.4) = y1 + Δy = 0.4227874, x2 = x1 + h = 0.4
o
Again, f(x, y) = 1 + y2, h = 0.2, x2 = 0.4, y2 = 0.4227874
.c
We have, k1 = h f (x2, y2) = 0.235749
FG h k IJ = 0.258463
k2 = h f x2 +H
F
n
= h f GH x
2
h
, y2 + 1
2
k I
K ts
+ J = 0.260945
2
2K
k3 2 + , y2
ra
2
k4 = h f (x2 + h, y2 + k3) = 0.293498
pi
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.261344
6 1
as
Thus, y3 = y(0.6) = y2 + Δy = 0.6841314.

Example 4. Use the Runge-Kutta fourth order method to find the value of y when x = 1
dy y − x
g
given that y = 1 when x = 0 (taking n = 2) and = . [U.P.T.U. (MCA) 2008]

dx y + x
.c
y− x
Sol. Here, f(x, y) = , x = 0, y0 = 1, h = 0.5
w
y+ x 0
We have, k1 = h f (x0, y0) = 0.5
w
FG h k
, y0 + 1
IJ = 0.333
H
k2 = h f x0 +
K
w
2 2
F
= h f Gx
h 2 k I
+ J = 0.3235
k3
H 0
2
, y0+
2K
k4 = h f (x0 + h, y0 + k3) = 0.2258
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.3398
6 1
Thus, y1 = y(0.5) = y0 + Δy = 1.3398, x1 = x0 + h = 0.5
Again, k1 = h f (x1, y1) = 0.22823
FG h k
, y1 + 1
IJ = 0.15969
H
k2 = h f x1 +
2 2 K
www.cgaspirants.com
www.cgaspirants.com
FG h k
, y1 + 2 = 0.15432
IJ
k3 = h f x1 +
H 2 2 K
k4 = h f (x1 + h, y1 + k3) = 0.09906
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.159218
6 1
Thus, y2 = y(1.0) = y1 + Δy = 1.499018.
Example 5. Use Runge-Kutta method of fourth order to approximate y when x = 0.1
dy
given that y = 1 at x = 0 and = 3x + y2. [U.P.T.U. MCA (C.O.) 2008]
dx
Sol. Here, f(x, y) = 3x + y2, h = 0.1, x0 = 0, y0 = 1
m
We have, k1 = h f (x0, y0) = 0.1
FG h k
, y0 + 1
IJ = 0.12525
k2 = h f x0 +H K
o
2 2
F k I
.c
h
= h f GH x + J = 0.127917
2
2K
k3 , y0
0 +
2
∴
k4 = h f (x0 + h, y0 + k3) = 0.15722
Δy =
1
n
(k + 2k2 + 2k3 + k4) = 0.127259
ts
6 1
ra
Thus, y(0.1) = y0 + Δy = 1.127259.

Example 6. Find the value of y(1.1) using Runge-Kutta method of fourth order, given
pi
dy
that = y2 + xy, y(1) = 1.0, take h = 0.05. (G.B.T.U. 2011, 2012; M.T.U. 2013)
dx
as
Sol. Here, f(x, y) = y2 + xy, x0 = 1, y0 = 1, h = 0.05

We have, k1 = h f (x0, y0) = 0.1
FG IJ = 0.10894
g
h k
, y0 + 1
k2 = h f x0 + H K
.c
2 2
F
= h f GH x
h k I
+ J = 0.109637
w
2
2K
k3 0 + , y0
2
w
k4 = h f (x0 + h, y0 + k3) = 0.119821

1
w
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.109496

6 1
Thus, y1 = y(1.05) = y0 + Δy = 1.109496, x1 = x0 + h = 1.05
Again, k1 = h f (x1, y1) = 0.119798
FG h k
, y1 + 1
IJ = 0.13123
H
k2 = h f x1 +
2 2 K
F
= h f GH x
h k I
+ J = 0.13221
2
2K
k3 1 + , y1
2
k4 = h f (x1 + h, y1 + k3) = 0.145385
www.cgaspirants.com
www.cgaspirants.com
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.13201
6 1
Thus, y2 = y(1.1) = y1 + Δy = 1.241506
Example 7. Use Runge-Kutta IV order method with h = 0.1 to find x(0.1) and x(0.2)
dx
where = t – x and x(0) = 0.
dt
Sol. Here, f(t, x) = t – x, t0 = 0, x0 = 0
k1 = hf (t0, x0) = 0
FG h k
, x0 + 1
IJ = 0.005
H
k2 = hf t0 +
2 2 K
m
F
= hf GH t
h k I
+ J = 0.00475
2
2K
k3 0 + , x0
o
2
k4 = hf (t0 + h, x0 + k3) = 0.009525
.c
1
∴ Δx = (k + 2k2 + 2k3 + k4) = 0.0048375
Thus,
Again,
6 1
ts
x(0.1) = x0 + Δx = 0.0048375, t1 = t0 + h = 0.1
n
k1 = hf (t1, x1) = 0.00951625
ra
FG h k IJ
, x1 + 1 = 0.014040
H
k2 = hf t1 +
2 2 K
pi
F
= hf G t
h k I
+ J = 0.013814
2
k3
H 1 +
2
, x1
2K
as
k4 = hf (t1 + h, x1 + k3) = 0.018135

1
g
∴ Δx = (k + 2k2 + 2k3 + k4) = 0.013893

6 1
.c
Thus, x(0.2) = x1 + Δx = 0.0187305.

Example 8. Solve the initial value problem u′ = – 2tu2, u(0) = 1 with h = 0.2 on the
w
interval [0, 0.4]. Use Runge-Kutta fourth order method and compare your result with exact
solution. [U.P.T.U. 2009, U.P.T.U. (MCA) 2006]
w
Sol. Here, f(t, u) = – 2t u2, t0 = 0, u0 = 1, h = 0.2

w
We have, k1 = h f (t0, u0) = 0
FG h k
, u0 + 1
IJ = – 0.04
H
k2 = hf t0 +
2 2 K
F
= hf G t
h k I
+ J = – 0.038416
2
k3
H 0 +
2
, u0
2K
k4 = hf (t0 + h, u0 + k3) = – 0.0739715
∴ Δu = – 0.03846725
Thus, u1 = u(0.2) = u0 + Δu = 0.9615328
www.cgaspirants.com
www.cgaspirants.com
Again, k1 = hf (t1, u1) = – 0.0739636
FG h k
, u1 + 1
IJ = – 0.1025753
H
k2 = hf t1 +
2 2 K
F
= hf G t
h k I
+ J = – 0.0994255
2
k3
H 1 +
2
, u1
2K
k4 = hf (t1 + h, u1 + k3) = – 0.1189166
∴ Δu = – 0.0994803
Thus, u2 = u(0.4) = u1 + Δu = 0.8620525
Absolute errors in numerical solutions are
m
∈(0.2) = | 0.961539 – 0.961533 | = 0.000006
o
∈(0.4) = | 0.862069 – 0.862053 | = 0.000016
Example 9. Using Runge-Kutta method of fourth order, find y(0.8) correct to 4 decimal
.c
places if y′ = y – x2, y(0.6) = 1.7379, taking h = 0.1. (U.P.T.U. 2007)
Sol. Here,
We have,
n
FG h k
ts
f(x, y) = y – x2, x0 = 0.6, h = 0.1, y0 = 1.7379
k1 = hf(x0, y0) = h(y0 – x02) = 0.13779
IJ = h LMFG y + k IJ − FG x + h IJ OP = 0.1384295
2
, y0 + 1
H
k2 = hf x0 +
K MNH 2 K H 2 K PQ 1
ra
0 0
2 2
FG k I LF k I F h I O
pi
2
+ J = h MG y + J − G x + J P
h 2 2
k3 = hf x0 +
H 2
, y0
2K MNH 2 K H 2 K PQ
0 0
as
= 0.138461475
k4 = hf (x0 + h, y0 + k3) = h[(y0 + k3) – (x0 + h)2] = 0.1386361475
g
1
Δy= (k + 2k2 + 2k3 + k4) = 0.138368
.c
∴
6 1
Thus, y1 = y(0.7) = y0 + Δy = 1.876268, x1 = x0 + h = 0.7
w
Now, x1 = 0.7, y1 = y(0.7) = 1.876268, h = 0.1

Again, k1 = hf(x1, y1) = h(y1 – x12) = 0.1386268
w
FG h k IJ LMFG y + k IJ − FG x + h IJ OP = 0.13830814
2
w
, y1 + 1 = h 1
k2 = hf x1 +
H 2 2 K MNH 2 K H 2 K PQ
1 1
F
= hf G x
h k I LF k I F h I O
+ J = h MG y + J − G x + J P
2 2
2
k3
H 1 +
2
, y1
2K MNH 2 K H 2 K PQ
1 1
= 0.138292207
k4 = hf(x1 + h, y1 + k3) = h[(y1 + k3) – (x1 + h)2] = 0.1374560207
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.138213919
6 1
Thus, y2 = y(0.8) = y1 + Δy = 2.0145, x2 = x1 + h = 0.8.
www.cgaspirants.com
www.cgaspirants.com
1
Example 10. Solve: y′ = , y(0) = 1 for x = 0.5 to x = 1 by Runge-Kutta method. Take
x+ y
h = 0.5.
1
Sol. Here, f(x, y) = , x = 0, h = 0.5, y0 = 1
x+ y 0
F 1 I = 0.5
We have, k1 = hf(x0, y0) = h GH x 0 + y0
JK
FG h k
, y0 + 1
IJ = h 1
H
k2 = hf x0 +
2 2 K FG x + h + y + k IJ = 0.3333 1
H 2 2K
0 0
m
FG h k I
+ J =h.
2 1
H
k3 = hf x0 +
2
, y0
2K FG x + h + y + k IJ
= 0.3529
o
2
H 2 2K
0 0
.c
1
k4 = hf (x0 + h, y0 + k3) = h = 0.2698
ts
( x0 + h + y0 + k3 )
1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.3570
n
6 1
Thus, y1 = y(0.5) = y0 + Δy = 1.3570, x1 = x0 + h = 0.5
ra
Now, x1 = 0.5, y1 = 1.3570, h = 0.5

h
Again, k1 = hf (x1, y1) = = 0.2692
pi
x1 + y1
FG h k
, y1 + 1 =
h IJ
as
k2 = hf x1 +
H 2 2 h
x1 + + y1 + 1
K k = 0.2230
2 2
F IJ
g
h k2 h
= hf G x
k3
H
1 + , y1 + =
h K k = 0.2254
.c
2 2 x1 + + y1 + 2
2 2
w
h
k4 = hf (x1 + h, y1 + k3) = = 0.1936
x1 + h + y1 + k3
w
1
∴ Δy = (k1 + 2k2 + 2k3 + k4) = 0.2266
6
w
Thus, y2 = y(1) = y1 + Δy = 1.5836, x2 = x1 + h = 1.

dy
Example 11. If = x + y2, use Runge-Kutta method of fouth order to find an approximate
dx
value of y for x = 0.2 given that y = 1 when x = 0. (Take h = 0.1) (U.P.T.U. 2009)
2
Sol. Here, f(x, y) = x + y , x0 = 0, y0 = 1, h = 0.1
We have, k1 = hf (x0, y0) = h(x0 + y02) = 0.1
FG h k IJ LM h k
, y0 + 1 = h x0 + + y0 + 1
FG IJ OP = 0.11525
2
H
k2 = hf x0 +
2 2 K MN 2 2 H K PQ
F
= hf G x
h k I L
+ J = hM x
2 h F
+ +Gy
k I O
2
+ J P = 0.116857
2
k3
H 0 +
2
, y0
2K MN 0
2 H
0
2 K PQ
www.cgaspirants.com
www.cgaspirants.com
k4 = hf (x0 + h, y0 + k3) = h[x0 + h + (y0 + k3)2] = 0.134737

1
∴ Δy = (k + 2k2 + 2k3 + k4) = 0.116492
6 1
Thus, y1 = y(0.1) = y0 + Δy = 1.116492, x1 = x0 + h = 0.1
Now, x1 = 0.1, y1 = 1.116492, h = 0.1
Again, k1 = hf (x1, y1) = h(x1 + y12) = 0.134655
FG h k
, y1 + 1
IJ = h LMx + h + FG y + k IJ OP = 0.155143 1
2
k2 = hf x1 +
H 2 2 K MN 2 H 2 K PQ 1 1
F
= hf G x
h k I L h F kI O
+ J = h M x + + G y + J P = 0.157579
2
m
2 2
k3
H 1 +
2
, y1
2K MN 2 H 2 K PQ 1 1
o
k4 = hf (x1 + h, y1 + k3) = h [x1 + h + (y1 + k3)2] = 0.1823257
.c
1
∴ Δy= (k + 2k2 + 2k3 + k4) = 0.15707
6 1
Thus, y2 = y(0.2) = y1 + Δy = 1.273562, x2 = x1 + h = 0.2.
Example 12. Solve:

dy
dx
= yz + x,
ndz
dx
ts
= xz + y ; given that y(0) = 1, z(0) = – 1 for y(0.1),
ra
z(0.1).
Sol. Here, f1(x, y, z) = yz + x, f2(x, y, z) = xz + y
pi
h = 0.1, x0 = 0, y0 = 1, z0 = – 1
k1 = hf1 (x0, y0, z0) = h (y0 z0 + x0) = – 0.1
as
l1 = hf2(x0, y0, z0) = h(x0 z0 + y0) = 0.1

FG h , y + k , z + l IJ = – 0.08525 1 1
H 2
k2 = hf1 x0 +
2K
g
0 0
2
.c
F h k , z + l IJ = 0.09025
l = hf G x + , y + 1 1
2 H 2 2 0
2 2K
0 0
w
F h k , z + l IJ = – 0.0864173
k = hf G x + , y + 2 2
H 2 2K
w
3 1 0 0 0
2
F h
l = hf G x + , y +
k l I
, z + J = 0.090963125
w
2 2
3
H 2 2 0
2 2K
0 0
k = hf b x + h, y + k , z + l g = – 0.073048
4 1 0 0 3 0 3
l4 = hf2(x0 + h, y0 + k3, z0 + l3) = 0.0822679

1
∴ k= (k + 2k2 + 2k3 + k4) = – 0.0860637
6 1
1
and l = (l1 + 2l2 + 2l3 + l4) = 0.0907823
6
Thus, y1 = y(0.1) = y0 + k = 1 – 0.0860637 = 0.9139363
z1 = z(0.1) = z0 + l = –1 + 0.0907823 = –0.9092176.
www.cgaspirants.com
www.cgaspirants.com
ASSIGNMENT
1. Use Runge-Kutta method to approximate y when x = 0.1 given that x = 0 when y = 1 and
dy
= x + y.
dx
dy
2. Use Runge-Kutta Fourth order formula to find y(1.4) if y (1) = 2 and = xy. Take h = 0.2.
dx
3. Use Runge-Kutta method to find y when x = 1.2 in steps of 0.1 given that
dy
= x2 + y2 and y(1) = 1.5
m
dx
4. (i) Write the main steps to be followed in using the Runge-Kutta method of fourth order to solve
o
dy
an ordinary diff. equation of first order. Hence solve = x3 + y 3, y(0) = 1 and step length
dx
.c
h = 0.1 up to three iterations.
(ii) State Runge-Kutta method of fourth order. Using this method, find the values of y(0.2), y(0.4)
and y(0.6) for the following initial value problem

n tsdy
dx
= x3 – y3 with condition that y(0) = 1.
(U.P.T.U. 2015)
ra
5. Use classical Runge-Kutta method of fourth order to find the numerical solution at x = 1.4 for
dy
pi
= y2 + x2, y(1) = 0. Assume step size h = 0.2.

dx
6. Given that y′ = x 2 – y, y(0) = 1, find y(0.1), y(0.2) using Runge-Kutta method of fourth order.
as
7. (i) Estimate y(1) if 2yy′ = x2 and y(0) = 2 using Runge-Kutta method of fourth order by taking
h = 0.5. Also compare the result with exact value. (G.B.T.U. 2011)
g
(ii) Estimate y (0.8) using Runge-Kutta method of fourth order (perform two iterations) for the
.c
dy
differential equation 2y = x2, y (0) = 2. Also compare the result with exact value.
dx
w
(G.B.T.U. 2013)
dy 1
w
8. Using Runge-Kutta method, find y(0.2) given that = 3x + y, y(0) = 1 taking h = 0.1.
dx 2
w
dy
9. Given that = 1 + xy; y(0) = 2. Using Runge-Kutta fourth order method, find y(0.1), y(0.2).
dx
[G.B.T.U. MCA (SUM) 2010]
10. Use Runge-Kutta method of fourth order to solve the following differential equation in the interval
[0, 0.4]:
dy y + x
= , y(0) = 1. Take h = 0.2. [G.B.T.U. (C.O.) 2011]
dx y − x
11. Using Runge-Kutta method of fourth order, solve the following differential equation:
dy y2 − x2
= with y (0) = 1 at x = 0.2, 0.4. (G.B.T.U. 2010)
dx y2 + x2
www.cgaspirants.com
www.cgaspirants.com
12. Use Runge-Kutta formula of fourth order to find the numerical solution at x = 0.6 and 0.8
for the differential equation y′ = x + y , y (0.4) = 0.41. Assume the step length h = 0.2.
[M.T.U. 2012, G.B.T.U. (MCA) 2011]
dy
13. (i) Apply Runge-Kutta fourth order method to solve 10 = x2 + y2 ; y (0) = 1 for 0 < x ≤ 0. 4 and
dx
h = 0.1. (G.B.T.U. 2013, 2011)
(ii) Apply Runge-Kutta method of fourth order to solve the initial value problem:
5 dy = x2 + y2, y (0) = 1
dx
and find y in the interval 0 ≤ x ≤ 0.2, taking h = 0.1. (G.B.T.U. 2012)
14. Using Runge-Kutta method of fourth order, solve for y(0.1), y(0.2) and y(0.3) given that
m
y′ = xy + y 2 , y(0) = 1. [M.T.U. (MCA) 2012, G.B.T.U. (C.O.) 2010]
2
Using fourth order Runge-Kutta method, solve the initial value problem d y − x dy + y2 = 0
o
15.
dx2 dx
.c
dy
with initial conditions y = 1 and = 0 when x = 0 in the interval [0, 0.2] and step size h = 0.1.
dx
16. Solve
dy
dx
= x + z,
dz
dx
n ts (M.T.U. 2013)
= x – y2 for y (0.1), z (0.1) given that y (0) = 2, z (0) = 1 by Runge-Kutta

ra
IV order method.
dy 2xy
17. Given the initial value problem = , y(1) = 3,
dx x 2 − y 2
pi
Find the numerical solution at x = 1.2 and x = 1.5 by using Runge-Kutta method of fourth order.
(U.P.T.U. 2014)
as
Answers
g
1. 1.11034 2. y(1.2) = 2.4921, y(1.4) = 3.2320

3. y(1.1) = 1.8955, y(1.2) = 2.5041
.c
4. (i) 1.118057, 1.291457, 1.584057 (ii) 0.8456, 0.7510, 0.7015

5. y(1.2) = 0.246326, y(1.4) = 0.622751489
w
6. y(0.1) = 0.9051627, y(0.2) = 0.8212695

w
7. (i) 2.0816705
(ii) y(0.4) = 2.005326, y(0.8) = 2.042221; y(0.8) [Exact value] = 2.042221829
w
8. 1.1749 9. 2.110359, 2.24309

10. y(0.2) = 1.23923, y(0.4) = 1.54892 11. 1.195999, 1.375269
12. y(0.6) = 0.61035, y (0.8) = 0.84899
13. (i) y(0.1) = 1.0101, y(0.2) = 1.0206, y(0.3) = 1.0317, y(0.4 ) = 1.0437
(ii) x: 0 0.1 0.2
y: 1 1.020475 1.04221
14. y(0.1) = 1.1168873, y(0.2) = 1.2773914, y(0.3) = 1.50412
15. x: 0 0.1 0.2
y: 1 0.995 0.98
16. 2.0845, 0.586.
17. y(1.2) = 2.8233
y(1.5) = 2.39265
www.cgaspirants.com
www.cgaspirants.com
TEST YOUR KNOWLEDGE
1. What do you mean by initial value problem? (U.P.T.U. 2014)

2. What do you mean by numerical differentiation? Explain in brief. (M.T.U. 2012)
3. Let I =
z
x0
x3
f ( x) dx where f(x) is a third degree polynomial. Write the formula you will like to use
to find the approximate value of I. It is given that the data are equispaced. (M.T.U. 2012)
4. Derive Newton-Cote’s quadrature formula for numerical integration. (M.T.U. 2013)
5. Show that y′ =
1 LM 1 1 1
Δy − Δ2 y + Δ3 y − Δ2 y + ... .
OP (M.T.U. 2013)
h N 2 3 4 Q
m
6. Calculate the value of z 5.2
log e x dx by trapezoidal rule. (M.T.U. 2013)
o
4
7. Explain two types of errors in numerical differentiation. [M.T.U. (MCA) 2012]
.c
8. Evaluate
z 6 dx
by using Simpson’s one-third rule. [M.T.U. (MCA) 2012]
9.
10.
0 1 + x2
ts
Discuss and explain the working of Runge-Kutta’s II and IV order methods.
What is the disadvantage of Picard’s method?
n
11. Given the table:
ra
x: 0 0.5 1
y: 1 0.8 0.5
z
pi
1
then find y dx by trapezoidal rule.
0
as
12. Evaluate
z 1
0
dx
1 + x2
(taking h = 1/4) by Simpson’s 1/3 rule.
g
13. Derive Picard’s method of successive approximations.

.c
14. Derive Simpson’s 1/3rd rule from Newton-Cote’s quadrature formula. Explain its limitations and
usefulness.
w
15. Derive Simpson’s 3/8th rule from Newton-Cote’s quadrature formula. (M.T.U. 2014)
16. Derive Trapezoidal rule from Newton-Cote’s quadrature formula.
w
17. What are the single-step and multi-step methods? Differentiate.

18. What do you mean by a boundary value problem?
w
19. Explain Milne’s Simpson predictor-corrector method for solution of an ordinary differential
equations numerically.
20. Write a short note on numerical differentiation.
Answers
3. Simpson’s 3/8th rule 6. 1.82765 8. 1.366173413 11. 0.775
12. 0.7854.
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
EXAMINATION PAPERS
U.P.T.U., LUCKNOW NAS–301
B. Tech. [SEMESTER-III]
(ODD SEM.) THEORY EXAMINATION, 2014–2015
MATHEMATICS –III
(PAPER ID: 199320)
Time: 3 Hours Total Marks: 100
m
Note: Attempt All Questions. All Questions carry equal marks.
o
1. Attempt any four parts of the following: (5 × 4 = 20)
.c
(a) State Cauchy-Riemann theorem for an analytic function. Test the analyticity of the
following function :
R| (x
f(z) = S
|T
3
2
x +y
0
2
n 3
– y ) + i (x + y )
3
, if z = 0
3
, if x ≠ 0 and ts
ra
(b) State Cauchy-integral theorem for an analytic function. Verify this theorem by
integrating the function z3 + iz along the boundary of the rectangle with vertices
pi
+ 1, – 1, i, – i.
1
(c) Show that the function u = log(x2 + y2) is harmonic. Find the harmonic conjugate
as
2
of u.
z
g
e2z
(d ) Evaluate the integral dz, around the boundary of the circle | z | = 2.
.c
( z + 1) 5
w
(e) Find the Taylor series expansion of the function tan–1 z about the point z = π/4.
z π cos 2 3θ
w
( f ) Evaluate the integral dθ .

0 5 – 4 cos 2θ
w
2. Attempt any two parts of the following: (10 × 2 = 20)

(a) Find the Fourier transform of the following function f(x) = 1 – x 2, if | x | ≤ 1 and
f(x) = 0, if |x| > 1.
(b) Using z-transform, solve the following difference equation
Yn+2 – (2 cos α)Yn + 1 + Yn = 7n with the conditions that Y0 = 5, Y1 = 1.
(c) State the convolution theorem for Fourier transform. Prove that the Fourier transform
of the convolution of the two functions is equal to the product of their Fourier
transforms.
(a) Define skewness and kurtosis of a distribution. The first four moments of a
distribution are 0, 2.5, 0.7, and 18.71. Find the coefficient of skewness and kurtosis.
555
www.cgaspirants.com
www.cgaspirants.com
(b) Fit a second degree parabola to the following data:
x 1 2 3 4 5 6 7 8 9
y 2 6 7 8 10 11 8 13 5
(c) Define coefficient of correlation and regression. If θ is the acute angle between the
1 – r2 σ xσ y
two lines of regression then prove that tan θ =
r σ 2x + σ 2y
where r, σx, σy have their usual meanings. Give the significance of the formula when
r = 0 and r = ± 1.
m
(a) Derive Newton-Raphson’s method to find a root of the equation f(x) = 0. Prove that
o
this method has quadratic convergence.
(b) Apply Newton’s divided difference method to obtain an interpolatory polynomial for
.c
the following data:
x
f(x)
3
31
5
51
n 7
17 ts 9
19
11
90
13
110
(c) Obtain Lagrange’s Interpolatory for the following data:

ra
x 1 3 5 7 10
pi
f(x) 13 31 25 37 101
Find the values of f(4) and f(8.5).

as

(a) Solve the following system of linear equations using Gauss-Seidel method
g
10x + 3y + 7z = 41
.c
3x + 20y + 17z = 101

x + 19y + 23z = 201
w
Perform three iterations.

(b) State Simpson’s three-eighth rule. Using this rule, evaluate the following integral
w
z 6 x
w
dx.
0 1 + x5
(c) State Runge-Kutta method of fourth order. Using this method, find the values of
dy
y(0.2), y(0.4) and y(0.6) for the following initial value problem = x3 – y3 with
dx
condition that y(0) = 1.
www.cgaspirants.com
www.cgaspirants.com
U.P.T.U., LUCKNOW AS–303
B. Tech. [SEMESTER-III]
(ODD SEM.) THEORY EXAMINATION, 2014–2015
ENGINEERING MATHEMATICS –III
(PAPER ID: 990303)
UNIT–1
m
1. Answer any four from the following: (4 × 5 = 20)
F∂ I
o
2
∂2
(a) If f(z) is a regular function of z, then prove that GH ∂x 2
+ 2 JK |f(z)|2 = 4 |f ′(z)|2.
.c
∂y
(b) Find the analytic function f(z) = u + iv, given that v = ex (x sin y + y cos y).
n
(c) Evaluate the following integral using Cauchy’s integral formula
ts z 4 – 3z
C z( z – 1)( z – 2)
dz
ra
3
where C is the circle |z| = .
2
pi
1
(d) Expand f(z) = for 1 < | z | < 2.
( z – 1) ( z – 2)
as
(e) Determine the poles of the following function and residue at each pole:
z
g
z2 z 2 dz
f(z) = and hence evaluate where C | z | = 3.
.c
( z – 1) 2 ( z + 2) C ( z – 1) 2 ( z + 2)
z
w
2π dθ
(f ) Evaluate 2 + cos θ by contour integration in the complex plane.
w
UNIT–2
w
e – ax
(a) Find Fourier sine transform of f(x) = .
x
(b) Using Parseval’s identity, show that z0

∝ x 2 dx
2
( x + 1) 2
=
π
4
.
∂u ∂ 2u
(c) Solve the equation = , x > 0, t > 0 subject to the condition:
∂t ∂x 2
557
www.cgaspirants.com
www.cgaspirants.com
(i) u = 0 when x = 0, t > 0 (ii) u =

RS1, 0 < x < 1 when t = 0
T0, x ≥ 1
(iii) u(x, t) is bounded.
(d) Solve the difference equation yk + 1 – 2yk–1 = 0, k ≥ 1, y(0) = 1.
(e) Find the Z-transform of sin αk, k ≥ 0.
9 z3
(f) Find Z–1 .
(3 z – 1) 2 ( z – 2)
(a) Three urns contains 6 red, 4 black; 4 red, 6 black, 5 red, 5 black balls respectively.
m
One of the urns is selected at random and a ball is drawn from it. If the ball drawn
is red, find the probability that it is drawn from the first urn.
o
(b) Using Poisson distribution, find the probability that the ace of spades will be drawn
from a pack of well-shuffled cards at least once in 104 consecutive trials.
.c
(c) Find the mean and standard deviation of Normal distribution.
(d) A manufacturer of envelopes knows that the weight of the envelopes is normally
ts
distributed with mean 1.9 gm and variance 0.01 square gm. Find how many envelopes
weighing (i) 2 gm or more, (ii) 2.1 gm or more, can be expected in a given packet of
1000 envelopes.
n
ra
[Given: If t is the normal variable, then φ(0 ≤ t ≤ 1) = 0.3413 and φ(0 ≤ t ≤ 2) = 0.4772]
(e) Find the moment generating function of Binomial distribution about its mean.
(f) If the probability density function of a random variable x is
pi
f(x) =
RSkx (1 – x)
α –1 β–1
, 0 < x < 1, α > 0, β > 0
as
T0, otherwise
Find k and mean of x.
g
4. Answer any two from the following: (2 × 10 = 20)

.c
(a) If an approximate root of the equation x(1 – loge x) = 0.5 lies between 0.1 and 0.2,
find the value of the root correct to three decimal places by Newton-Raphson method.
w
(b) Solve the system of equations

x + y + 54z = 110
w
27x + 6y – z = 85
6x + 15y + 2z = 72
w
Using Gauss-Seidel iteration method.

(c) Find the cubic spline approximation for the function y = f(x) from the following data,
given that y0 = y3 = 0
x –1 0 1 2
y –1 1 3 35
www.cgaspirants.com
www.cgaspirants.com
EXAMINATION PAPERS 559
5. Answer any two from the following: (2 × 10 = 20)

(a) The velocity V of a particle at distances from a point on its path is given by the table:
S 0 10 20 30 40 50 60 feet
V 47 58 64 65 61 52 38 feet/sec
Estimate the time taken to travel 60 feet by using Simpson’s one-third rule. Compare
3
the result with Simpson’s rule.
8
(b) By applying the fourth order Runge-Kutta Method find y(0.2) from y′ = y – x, y(0) = 2
taking h = 0.1
m
dy
(c) The differential equation = y – x2 is satisfied by y(0) = 1, y(0.2) = 1.12186,
dx
o
y(0.4) = 1.46820, y(0.6) = 1.7379. Compute the value of y(0.8) by Milne’s predictor-
.c
corrector formula.
n ts
ra
pi
g as
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
U.P.T.U., LUCKNOW
B. Tech. (SEM. III) ODD SEMESTER THEORY AS–303
EXAMINATION 2013–2014
MATHEMATICS –III
(PAPER ID: 1226)
Note: Attempt all questions from each Section as indicated. The symbols have their usual meaning.
SECTION–A
m
1. Attempt all parts of this Section. Each part carries 2 marks: (2 × 10 = 20)
o
(a) Define Conformal Transformation.
.c
z2
(b) Find residue of f(z) = at the pole – 1.
z2 + 3 z + 2
(c)
(d)
Define Fourier Transform of a function f(x).
n
Find the Z-Transform of {ak}, k ≥ 0.
ts
(e) Define coefficients of Skewniess.
ra
(f) What is Total Probability Theorem?

(g) Define Spline Function.
pi
1 1
–
(h) Show that δ = E 2 – E 2 .
as
(i) Define rate of convergence.

(j) What do you mean by initial value problem?
g
.c
SECTION–B
w
2. Attempt any three parts of this Section. (10 × 3 = 30)
z z2 + 1
w
(a) State and prove Cauchy integral formula. Also evaluate dz, where C is the
C z2 – 1
w
circle:
1
(i) | z – 1 | = 1 (ii) | z | = .
2
∂y ∂ 2 y
(b) Using Fourier Transform, solve = , – ∞ < x < ∞, t > 0; y(x, 0) = f(x).
∂t ∂x 2
(c) The first four moments of a distribution about the value ‘4’ of the variables are – 1.5,
17, – 30 and 108. Find the moments about mean and about origin. Also find Skewness
and Kurtosis.
560
www.cgaspirants.com
www.cgaspirants.com
EXAMINATION PAPERS 561
(d) Use Gauss-Seidel method to solve the following system of simultaneous equations:
9x + 4y + z = – 17
x – 2y – 6z = 14
x + 6y = 4
Perform four iterations.
dy
(e) Given = y – x, y(0) = 2. Find y(0.1) and y(0.2) correct to four decimal places by
dx
Runge-Kutta fourth order method.
SECTION–C
m
Note: All questions of this Section are compulsory. Attempt any two part from each question:
(5 × 2 × 5 = 50)
o
3. (a) Verify Cauchy’s theorem by integrating z3 along the boundary of a square with vertices
at 1 + i, 1 – i, – 1 + i and – 1 – i.
.c
ts
(b) Evaluate the following integral by using complex integration z
0
π cos 2θ
5 + 4 cos θ dθ.
(c) Determine the analytic function f(z) = u + iv, in terms of z, whose real part is
n
e–x (x sin y – y cos y).
ra
4. (a) Find the Fourier transform of: F(x) =

RS1,| x|< a. Hence evaluate
T0,| x|> a
pi
(i)
z ∞ sin ap cos px
dp (ii) z ∞ sin p
dp.
as
–∞ p 0 p
1
g
(b) Find the inverse Z-transform of: F(z) = ( z – 3) ( z – 2) for

.c
(i) |z| < 2 (ii) 2 < |z|< 3 (iii) |z| > 3.

(c) Solve by Z-transform the difference equation:
w
yk + 2 + 6yk+ 1 + 9yk = 2k; (y0 = y1 = 0).

w
5. (a) State and prove Baye’s theorem.

(b) A continuous random variable X has a p.d.f. f(x) = 3x2, 0 ≤ x ≤ 1. Find a and b such
w
that:
(i) P(x ≤ a) = P(x > a), and (ii) P(x > b) = 0.05.
(c) Fit a Poisson distribution to the following data and calculate theoretical frequencies:
Death 0 1 2 3
Frequencies 122 260 15 2
6. (a) Find a positive value of (17)1/3 correct to four decimal places by Newton-Raphson
method.
www.cgaspirants.com
www.cgaspirants.com
(b) Obtain cubic spline for the following data:

x 0 1 2 3
f(x) 1 2 32 244
With the end conditions M0 = M3 = 0 for (0, 1). Hence compute f(0.5).
(c) From the following table of half-yearly premium for policies maturing at different
ages, estimate the premium for policies maturing at age of 46.
Age 45 50 55 60 65
Premium (in rupees) 114.84 96.16 83.32 74.48 68.48
m
7. (a) The table given below reveals the velocity ‘v’ of a body during the time ‘t’ specified.
Find its acceleration at t = 1.1.
o
t 1.0 1.1 1.2 1.3 1.4
.c
v 43.1 47.7 52.1 56.4 60.8
(b) Evaluate z
0
6 ex
x +1
dx by Simpson’s 3/8th rule.
n ts
ra
dy
(c) Using Milne’s method, solve = 1 + y2 with initial condition y(0) = 0, y(0.2) =
dx
pi
0.2027, y(0.4) = 0.4228, y(0.6) = 0.6841, obtain y(0.8).

g as
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
B. Tech. [SEMESTER III/IV]

THEORY EXAMINATION, 2013–2014
MATHEMATICS –III
Note: Attempt all questions.
1. Attempt any four parts of the question: (5 × 4 = 20)

3
(a) Define analytic function. Discuss the analyticity of f(z) = Re(z ) in the complex plane.
(b) Show that v(x, y) = e–x (x cos y + y sin y) is harmonic. Find its harmonic conjugate.
m
(c) Integrate f (z) = Re(z) from z = 0 to z = 1 + 2i.
(i) along straight line joining z = 0 to z = 1 + 2i.
o
(ii) along the real axis from z = 0 to z = 1 and then along a line parallel to imaginary
axis from z = 1 to z = 1 + 2i.
.c
(d ) Evaluate z
C
(1 + z) sin z
(2 z – 3) 2
n ts
dz, where C is the circle|z – i| = 2 counter-clockwise.
(e) Find all Taylor and Laurent’s series expansion of the following function about z = 0:
– 2z + 3
ra
f(z) = 2
z – 3z + 2
z
pi
2π dθ
( f ) Use contour integration to evaluate .
0 3 – 2 cos θ + sin θ
as

(a) Define moment generating function. Why is it called moment generating function?
g
1
If P(X = x) = , x = 1, 2, 3, ... find the moment generating function of x. Hence obtain
.c
2x
the variance.
w
(b) Determine the normal equation if the curve y = ax + bx2 is fitted to the data (xi, yi),
i = 1, 2, ..., m. Hence fit this curve to the data:
w
x: 1 2 3 4 5
w
y: 1.8 5.1 8.9 14.1 19.8

(c) Calculate the coefficient of correlation between the following ages of husband (x)
and wife (y) by taking 30 and 28 as assumed mean in case of x and y respectively:
x: 24 27 28 28 29 30 32 33 35 35 40
y: 18 20 22 25 22 28 28 30 27 30 32
(a) Out of 800 families with four children each, how many families would be expected to
have
(i) 2 boys and 2 girls (ii) at least one boy
(iii) no girl (iv) atmost two girls.
Assume equal probabilities for boys and girls.
563
www.cgaspirants.com
www.cgaspirants.com
(b) The groups of 100 people each were taken for testing the use of a vaccine. 15 persons
contracted the disease out of the inoculated persons, while 25 contracted the disease
in the other group. Test the efficiency of the vaccine using chi-square test. (the value
of χ2 for one degree of freedom at 5% level of significance is 3.84)
(c) Calculate the trend values by the method of least square fit to a straight line and
hence estimate profit for 1981:
Year : 1971 1972 1973 1974 1975 1976 1977
Profit (in thousands) : 60 72 75 65 80 85 95
4. Attempt any four parts of the following:
(a) Explain Newton-Raphson method and use it to find the positive root of x4 = x + 10
correct to three decimal places.
m
(b) Find the root of the equation 2x (1 – x2 + x) ln x = x2 – 1 lying in the interval [0, 1]
using Regula-Falsi method.
o
(c) Prove that 1 + δ2μ2 = 1 +
FG 1 2 IJ 2
H δ
K , where symbols have their usual meanings for
.c
2
finite differences.
data:
x: 21 25
n
29 33
ts
(d) Use Newton-Gregory formula to interpolate the value of y at x = 36 from the following
37
y: 18.4 17.8 17.1 16.3 15.5
ra
(e) Find f (x) as a polynomial in x for the following data using Newton’s divided difference
formula:
pi
x: –4 –1 0 2 5
f(x) : 1245 33 5 9 1335
as
(f) Using Lagrange’s interpolation formula, find polynomial which takes the values 3,
12, 15, – 21 when x has the values 3, 2, 1, – 1.
5. Attempt any two parts of the following:
g
4 1 LM
1 OP
.c
(a) Decompose A = 1 4 – 2 in the form LU, where L is lower triangular matrix and
3 2 –4 MN PQ
w
U is the upper triangular matrix and hence solve the system of equations:
4x1 + x2 + x3 = 6
w
x1 + 4x2 – 2x3 = 4
w
3x1 + 2x2 – 4x3 = 6

(b) (i) A slider in a machine moves along fixed straight rod. Its distances x(m) along
the rod are given at various times (sec.)
t: 1 1.1 1.2 1.3 1.4 1.5
x: 16.40 19.01 21.96 25.29 29.03 33.21
Find the velocity of the slider at t = 1.1 sec.
(ii) Evaluate
z
0
2π
(e – t sin 10t) dt using Simpson’s rule with eight intervals.
dy 2 xy
(c) Given the initial value problem = 2 , y(1) = 3
dx x – y2
Find the numerical solution of x = 1.2 and x = 1.5 by using Runge-Kutta method of
fourth order.
www.cgaspirants.com
www.cgaspirants.com
B. Tech. (SEM. III) ODD SEMESTER THEORY AS301

EXAMINATION, 2013–2014
MATHEMATICS –III
Note: Attempt all questions from each Section as indicated. The symbols have their usual meaning.
SECTION–A
1. Attempt all parts of this Section. Each part carries 2 marks: (2 × 10 = 20)
2z + 1
m
(a) Find residue of f(z) = 2
at the pole z = – 1.
z – z–2
(b) Define harmonic function.
o
(c) State Convolution theorem for Fourier Transform.
.c
(d) Find the Z-Transform of {nCk}, 0 ≤ k ≤ n.
ts
(e) Define coefficients of kurtosis.
(f ) Define marginal and conditional distribution.
n
(g) Prove that: |(X, Y)| < ||X|| ||Y||.
(h) Define Abelian group.
ra
(i) Define rate of convergence.

(j) Write the formula for Simpson’s 3/8 rule.
pi
SECTION–B
as
Note: Attempt any three parts of this Section. (10 × 3 = 30)
z ∞ cosh ax 1 a
g
2. (a) Apply calculus residues to prove that: dx = sec .

0 cosh πx 2 2
.c
1 2
w
(b) If Fc(p) = tan–1 2 , then find f(x).

2 p
w
(c) Show that Poisson distribution is a limiting form of binomial distribution when p is
a very small and n is very large. Also find mean and variance of Poisson distribution.
w
(d) If p = p(x) = p0 + p1x + p2x2 and q = q(x) = q0 + q1x + q2x2, then the inner product is
defined by:
(p, q) = p0q0 + p1q1 + p2q2 for the vectors X1 = 1 + 2x + 3x2, X2 = 3 + 5x + 5x2,
X3 = 2 + x + 8x2. Find the orthogonal vectors.
(e) Use Gauss-Seidel method to solve the following system of simultaneous equations:
83x + 11y – 4z = 95
7x + 52y + 13z = 104
3x + 8y + 29z = 71
Perform four iterations.
565
www.cgaspirants.com
www.cgaspirants.com
SECTION–C
Note: All questions of this Section are compulsory. Attempt any two parts from each question:
(10 × 5 = 50)
3. (a) In a two-dimensional fluid flow, the stream function is
y
ψ=– 2 , find the velocity potential Φ.
x + y2
z
(b) Expand f(z) = in Laurent series vaild for region:
( z – 1) ( z – 2)
(i) | z – 1 | > 1 (ii) 0 < | z – 2 | < 1.
(c) State and prove Cauchy’s Theorem.
m
1 x
4. (a) Find Fourier cosine transform of and hence find Fourier sine Transform of .
1 + x2 1 + x2
o
.c
9 z3
(b) Find the inverse Z-transform of F(z) = .
( z – 2) (3 z – 1) 3
(c) Solve by Z-transform the difference equation:
yk + 2 – 2yk + 1 + yk = 3k + 5, y(0) = 0, y(1) = 1.
n ts
5. (a) In a certain factory manufacturing razor blades, there is a small chance of 0.002 for
ra
any blade to be defective. The blades are supplied in packets of 10. Use suitable
distribution to calculate the approximate number of packets containing no defective,
one defective and two defective blades respectively in a consignment of 20,000 packets.
pi
(b) Find the moment generating function of the exponential distribution:

1 –x/c
as
f(x) = e , 0 ≤ x ≤ ∞, c > 0. Hence find its mean and S.D.

c
(c) Calculate the first four moments about the mean for the following data:
g
Class-interval: 0–10 10–20 20–30 30–40 40–50

.c
Frequency: 10 20 40 20 10
6. (a) Examine the following vectors for linear dependence and find the relation, if it exists:
w
X1 = (1, 2, 1), X2 = (3, 1, 5), X3 = (3, – 4, 5).

(b) Let V be the vector space of all real valued continuous functions over R. Then show
w
that the solutions set W of the differential equation:

d2 y dy
w
3 2
+ 11– 4y = 0, is a subspace of V.
dx dx
(c) Show that the intersection of any two subspaces of a vector space is also a space of
the same.
7. (a) Use Newton’s Divided difference formula to find f(x) from the following data:
x: 0 1 2 4 5 6
f(x): 1 14 15 5 6 19
(b) Compute the rate of convergence of Newton-Raphson method.
(c) Apply Runge-Kutta fourth order method to find an approximate value of y when
dy
x = 0.2, given that = x + y with initial condition y = 1 at x = 0.
dx
www.cgaspirants.com
www.cgaspirants.com
APPENDIX
Table 1: NORMAL TABLE

AREAS UNDER THE STANDARD NORMAL
CURVE =
1
2π z0
z
e
−
z2
2 dz
m
z=0 z
o
z .00 .01 .02 .03 .04 .05 .06 .07 .08 .09
.c
0.0 .0000 .0040 .0080 .0120 .0160 .0199 .0239 .0279 .0319 .0359
0.1 .0398 .0438 .0478 .0517 .0557 .0596 .0636 .0675 .0714 .0754
0.2 .0793 .0832 .0871 .0910 .0948 .0987 .1026 .1064 .1103 .1141
0.3
0.4
0.5
.1179
.1554
.1915
.1217
.1591
.1950
.1255
.1628
.1985
.1293
.1664
.2019
n .1331
.1700
.2054
.1368
.1736
.2088 ts
.1406
.1772
.2123
.1443
.1808
.2157
.1480
.1844
.2190
.1517
.1879
.2224
0.6 .2257 .2291 .2324 .2357 .2389 .2422 .2454 .2485 .2517 .2549
ra
0.7 .2580 .2611 .2642 .2673 .2704 .2734 .2764 .2794 .2823 .2852
0.8 .2881 .2910 .2939 .2967 .2995 .3023 .3051 .3078 .3106 .3133
0.9 .3159 .3186 .3212 .3238 .3264 .3289 .3315 .3340 .3365 .3389
pi
1.0 .3413 .3438 .3461 .3485 .3508 .3531 .3554 .3577 .3599 .3621
1.1 .3643 .3665 .3686 .3708 .8729 .3749 .3770 .3790 .3810 .3830
as
1.2 .3849 .3869 .3888 .3907 .3925 .3944 .3962 .3980 .3997 .4015
1.3 .4032 .4049 .4066 .4082 .4099 .4115 .4131 .4147 .4162 .4177
1.4 .4192 .4207 .4222 .4236 .4251 .4255 .4279 .4292 .4306 .4319
g
1.5 .4332 .4345 .4357 .4370 .4382 .4394 .4406 .4418 .4429 .4441
.c
1.6 .4452 .4463 .4474 .4484 .4495 .4505 .4515 .4525 .4535 .4545
1.7 .4554 .4564 .4573 .4582 .4591 .4599 .4608 .4616 .4625 .4633
w
1.8 .4641 .4649 .4656 .4664 .4671 .4678 .4686 .4693 .4699 .4706
1.9 .4713 .4719 .4726 .4732 .4738 .4744 .4750 .4756 .4761 .4767
2.0 .4772 .4778 .4783 .4788 .4793 .4798 .4803 .4808 .4812 .4817
w
2.1 .4821 .4826 .4830 .4834 .4838 .4842 .4846 .4850 .4854 .4857
2.2 .4861 .4864 .4868 .4871 .4875 .4878 .4881 .4884 .4887 .4890
w
2.3 .4893 .4896 .4898 .4901 .4904 .4906 .4909 .4911 .4913 .4916
2.4 .4918 .4920 .4922 .4925 .4927 .4929 .4931 .4932 .4934 .4936
2.5 .4938 .4940 .4941 .4943 .4945 .4946 .4948 .4949 .4951 .4952
2.6 .4953 .4955 .4956 .4957 .4959 .4930 .4961 .4962 .4963 .4964
2.7 .4965 .4966 .4967 .4968 .4969 .4970 .4971 .4972 .4973 .4974
2.8 .4974 .4975 .4976 .4977 .4977 .4978 .4979 .4979 .4980 .4981
2.9 .4981 .4982 .4982 .4983 .4984 .4984 .4985 .4985 .4986 .4986
3.0 .4987 .4987 .4987 .4988 .4988 .4989 .4999 .4989 .4990 .4990
3.1 .4990 .4991 .4991 .4991 .4992 .4992 .4992 .4992 .4993 .4993
3.2 .4993 .4993 .4994 .4994 .4994 .4994 .4994 .4995 .4995 .4995
3.3 .4995 .4995 .4995 .4996 .4996 .4996 .4996 .4996 .4996 .4997
3.4 .4997 .4997 .4997 .4997 .4997 .4997 .4997 .4997 .4997 .4998
3.5 .4998 .4998 .4998 .4998 .4998 .4998 .4998 .4998 .4998 .4998
(i)
www.cgaspirants.com
www.cgaspirants.com
(ii) A TEXTBOOK OF ENGINEERING MATHEMATICS
Table 2 : SIGNIFICANT VALUES tv (α) OF t-DISTRIBUTION

(TWO TAIL AREAS) [| t | > tv(α)] = α
d.f. Probability (Level of Significance)

(ν) 0.50 0.10 0.05 0.02 0.01 0.001
1 1.00 6.31 12.71 31.82 63.66 636.62
2 0.82 0.92 4.30 6.97 6.93 31.60
3 0.77 2.32 3.18 4.54 5.84 12.94
4 0.74 2.13 2.78 3.75 4.60 8.61
5 0.73 2.02 2.57 3.37 4.03 6.86
m
6 0.72 1.94 2.45 3.14 3.71 5.96
7 0.71 1.90 2.37 3.00 3.50 5.41
o
8 0.71 1.80 2.31 2.90 3.36 5.04
.c
9 0.70 1.83 2.26 2.82 3.25 4.78
10 0.70 1.81 2.23 2.76 3.17 4.59
11
12
13
0.70
0.70
0.69
1.80
1.78
1.77
n 2.20
2.18
2.16
ts 2.72
2.68
2.05
3.11
3.06
3.01
4.44
4.32
4.22
ra
14 0.69 1.76 2.15 2.62 2.98 4.14
15 0.69 1.75 2.13 2.60 2.95 4.07
pi
16 0.69 1.75 2.12 2.58 2.92 4.02

17 0.69 1.74 2.11 2.57 2.90 3.97
as
18 0.69 1.73 2.10 2.55 2.88 3.92

19 0.69 1.73 2.09 2.54 2.86 3.88
20 0.69 1.73 2.09 2.53 2.85 3.85
g
.c
21 0.69 1.72 2.08 2.52 2.83 3.83

22 0.69 1.72 2.07 2.51 2.42 3.79
w
23 0.69 1.71 2.07 2.50 2.81 3.77

24 0.69 1.71 2.06 2.49 2.80 3.75
w
25 0.68 1.71 2.06 2.49 2.79 3.73
26 0.68 1.71 2.06 2.48 2.78 3.71

w
27 0.68 1.70 2.05 2.47 2.77 3.69

28 0.68 1.70 2.05 2.47 2.76 3.67
29 0.68 1.70 2.05 2.46 2.76 3.66
30 0.68 1.70 2.04 2.46 2.75 3.65
∞ 0.67 1.65 1.96 2.33 2.58 3.29
www.cgaspirants.com
www.cgaspirants.com
APPENDIX (iii)
Table 3 : F-Distribution
Values of F for F-Distributions with 0.05 of the Area in the Right Tail
Degrees of freedom for numerator
1 2 3 4 5 6 7 8 9 10 12 15 20 24 30 40 60 120 ∞
1 161 200 216 225 230 234 237 239 241 242 244 246 248 249 250 251 252 253 254
2 18.5 19.0 19.2 19.2 19.3 19.3 19.4 19.4 19.4 19.4 19.4 19.4 19.4 19.5 19.5 19.5 19.5 19.5 19.5
3 10.1 9.55 9.28 9.12 9.01 9.94 8.89 8.85 8.81 8.79 8.74 8.70 8.66 8.64 8.62 8.59 8.57 8.55 8.53
4 7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00 5.96 5.91 5.86 5.80 5.77 5.75 5.72 5.69 5.66 5.63
m
5 6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77 4.74 4.68 4.62 4.56 4.53 4.50 4.46 4.43 4.40 4.37
6 5.99 5.14 4.76 4.53 4.39 4.28 4.21 4.15 4.10 4.06 4.00 3.94 3.87 3.84 3.81 3.77 3.74 3.70 3.67
o
7 5.59 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68 3.64 3.57 3.51 3.44 3.41 3.38 3.34 3.30 3.27 3.23
.c
8 5.32 4.46 4.07 3.84 3.69 3.58 3.50 3.44 3.39 3.35 3.28 3.22 3.15 3.12 3.08 3.04 3.01 2.97 2.93
9 5.12 4.26 3.86 3.63 3.48 3.37 3.29 3.23 3.18 3.14 3.07 3.01 2.94 2.90 2.86 2.83 2.79 2.75 2.71
10 4.96 4.10
11 4.84 3.98
3.71
3.59
3.48 3.33
3.36 3.20
3.22 3.14 3.07 3.02
n
3.09 3.01 2.95 2.90
2.98
2.85
ts
2.91
2.79
2.85 2.77 2.74 2.70 2.66 2.62 2.58 2.54
2.72 2.65 2.61 2.57 2.53 2.49 2.45 2.40
ra
12 4.75 3.89 3.49 3.26 3.11 3.00 2.91 2.85 2.80 2.75 2.69 2.62 2.54 2.51 2.47 2.43 2.38 2.34 2.30
13 4.67 3.81 3.41 3.18 3.03 2.92 2.83 2.77 2.71 2.67 2.60 2.53 2.46 2.42 2.38 2.34 2.30 2.25 2.21
pi
14 4.60 3.74 3.34 3.11 3.96 2.85 2.76 2.70 2.65 2.60 2.53 2.46 2.39 2.35 2.31 2.27 2.22 2.18 2.13
15 4.54 3.68 3.29 3.06 3.90 2.79 2.71 2.64 2.59 2.54 2.48 2.40 2.33 2.29 2.25 2.20 2.16 2.11 2.07
as
16 4.49 3.63 3.24 3.01 2.85 2.74 2.66 2.59 2.54 2.49 2.42 2.35 2.28 2.24 2.19 2.15 2.11 2.06 2.01
17 4.45 3.59 3.20 2.96 2.81 2.70 2.61 2.55 2.49 2.45 2.38 2.31 2.23 2.19 2.15 2.10 2.06 2.01 1.96
g
18 4.41 3.55 3.16 2.93 2.77 2.66 2.58 2.51 2.46 2.41 2.34 2.27 2.19 2.15 2.11 2.06 2.02 1.97 1.92
.c
19 4.38 3.52 3.13 2.90 2.74 2.63 2.54 2.48 2.42 2.38 2.31 2.23 2.16 2.11 2.07 2.03 1.98 1.93 1.88
20 4.35 3.49 3.10 2.87 2.17 2.60 2.51 2.45 2.39 2.35 2.28 2.20 2.12 2.08 2.04 1.99 1.95 1.90 1.84
w
21 4.32 3.47 3.07 2.84 2.68 2.57 2.49 2.42 2.37 2.32 2.25 2.18 2.10 2.05 2.01 1.96 1.92 1.87 1.81
w
22 4.30 3.44 3.05 2.82 2.66 2.55 2.46 2.40 2.34 2.30 2.23 2.15 2.07 2.03 1.98 1.94 1.89 1.84 1.78
23 4.28 3.42 3.03 2.80 2.64 2.53 2.44 2.37 2.32 2.27 2.20 2.13 2.05 2.01 1.96 1.91 1.86 1.81 1.76
w
24 4.26 3.40 3.01 2.78 2.62 2.51 2.42 2.36 2.30 2.25 2.18 2.11 2.03 1.98 1.94 1.98 1.84 1.79 1.73
25 4.24 3.39 2.99 2.76 2.60 2.94 2.40 2.34 2.28 2.24 2.16 2.29 2.01 1.96 1.92 1.87 1.82 1.77 1.71
30 4.17 3.32 2.92 2.69 2.53 2.42 2.33 2.27 2.21 2.16 2.09 2.01 1.93 1.89 1.84 1.79 1.74 1.64 1.62
40 4.08 3.23 2.84 2.61 2.45 2.34 2.25 2.18 2.12 2.08 2.00 1.92 1.84 1.79 1.74 1.69 1.64 1.58 1.51
60 4.00 3.15 2.76 2.53 2.37 2.25 2.17 2.10 2.04 1.99 1.92 1.84 1.75 1.70 1.65 1.59 1.53 1.47 1.39
120 3.92 3.07 2.68 2.45 2.29 2.18 2.09 2.02 1.96 1.91 1.83 1.75 1.66 1.61 1.55 1.50 1.43 1.35 1.25
∞ 3.84 3.00 2.60 2.37 2.21 2.10 2.01 1.94 1.88 1.83 1.75 1.67 1.57 1.52 1.46 1.39 1.32 1.22 1.00
www.cgaspirants.com
www.cgaspirants.com
(iv) A TEXTBOOK OF ENGINEERING MATHEMATICS
Table 4 : CHI-SQUARE
Significant Values χ2 (α) of Chi-Square Distribution Right Tail Areas
for Given Probability α,
P = Pr (χ2 > χ2 (α)) = α
And v is Degrees of Freedom (d.f.)
Degree of Probability (Level of Significance)

freedom (ν)
0.99 0.95 0.50 0.10 0.05 0.02 0.01
1 .000157 .00393 .455 2.706 3.841 5.214 6.635
m
2 .0201 .103 1.386 4.605 5.991 7.824 9.210
3 .115 .352 2.366 6.251 7.815 9.837 11.341
o
4 .297 .711 3.357 7.779 9.488 11.668 13.277
5 .554 1.145 4.351 9.236 11.070 13.388 15.086
.c
6 .872 2.635 5.348 10.645 12.592 15.033 16.812
ts
7 1.239 2.167 6.346 12.017 14.067 16.622 18.475
8 1.646 2.733 7.344 13.362 15.507 18.168 20.090
n
9 2.088 3.325 8.343 14.684 16.919 19.679 21.669
10 2.558 3.940 9.340 15.987 18.307 21.161 23.209
ra
11 3.053 4.575 10.341 17.275 19.675 22.618 24.725

12 3.571 5.226 11.340 18.549 21.026 24.054 26.217
pi
13 4.107 5.892 12.340 19.812 22.362 25.472 27.688

14 4.660 6.571 13.339 21.064 23.685 26.873 29.141
as
15 4.229 7.261 14.339 22.307 24.996 28.259 30.578

16 5.812 7.962 15.338 23.542 26.296 29.633 32.000
17 6.408 8.672 15.338 24.769 27.587 30.995 33.409
g
18 7.015 9.390 17.338 25.989 28.869 32.346 34.805

.c
19 7.633 10.117 18.338 27.204 30.144 33.687 36.191

20 8.260 10.851 19.337 28.412 31.410 35.020 37.566
w
21 8.897 11.591 20.337 29.615 32.671 36.343 38.932

w
22 9.542 12.338 21.337 30.813 33.924 37.659 40.289

23 10.196 13.091 22.337 32.007 35.172 38.968 41.638
w
24 10.856 13.848 23.337 32.196 36.415 40.270 42.980

25 11.524 14.611 24.337 34.382 37.65 41.566 44.314
26 12.198 15.379 25.336 35.363 38.885 41.856 45.642
27 12.879 16.151 26.336 36.741 40.113 41.140 46.963
28 13.565 16.928 27.336 37.916 41.337 45.419 48.278
29 14.256 17.708 28.336 39.087 42.557 46.693 49.588
30 14.933 18.493 29.336 40.256 43.773 47.962 50.892
Note. For degrees of freedom (ν) greater than 30, the quantity 2χ2 − 2ν − 1 may be used as a normal
variate with unit variance.
www.cgaspirants.com
www.cgaspirants.com
om
.c
n ts
ra
pi
gas
.c
w
w
w
www.cgaspirants.com
www.cgaspirants.com
o m
.c
ts
an
ir
sp
ga
.c
w
w
w
ISBN 978-93-80386-53-9
9789380386539- 0475
(An ISO 9001:2015 Company)
www.cgaspirants.com

(N.P. Bali, Manish Goyal) A Textbook of Engineerin PDF

Uploaded by

Copyright:

Available Formats

(N.P. Bali, Manish Goyal) A Textbook of Engineerin PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(N.P. Bali, Manish Goyal) A Textbook of Engineerin PDF

Uploaded by

Copyright:

Available Formats

What is the textbook about?

What is the textbook about?

What are the different editions of the textbook?

What are the different editions of the textbook?

www.cgaspirants.

N.P. BALI Dr. MANISH GOYAL

S.B. College, Gurgaon Associate Professor

Haryana Department of Mathematics

G.L.A. University, Mathura

UNIVERSITY SCIENCE PRESS

A TEXTBOOK OF ENGINEERING MATHEMATICS

© by Laxmi Publications (P) Ltd.

Printed and bound in India

& Bengaluru 080-26 75 69 30

& Chennai 044-24 34 47 26, 24 35 95 07

& Guwahati 0361-254 36 69, 251 38 81

& Hyderabad 040-27 55 53 83, 27 55 53 93

& Jalandhar 0181-222 12 72

& Kolkata 033-22 27 43 84

UNIVERSITY SCIENCE PRESS

An ISO 9001:2008 Company

Syllabus ... (x)

Standard Results ... (xi)–(xv)

PREFACE TO THE EIGHTH EDITION

eÕ;kos'; euks ;s eka fuR;;qDrk miklrsA

its application to solve difference equations.

Unit III: Statistical Techniques 8

Moments, Moment generating functions, Skewness, Kurtosis, Curve fitting, Method of

Unit IV: Numerical Techniques–I 8

Zeroes of transcendental and polynomial equations using Bisection method, Regula-

falsi method and Newton-Raphson method, Rate of convergence of above methods.

Lagrange’s and Newton’s divided difference formula for unequal intervals.

Unit V: Numerical Techniques–II 8

21. cosh2 x – sinh2 x = 1, sech2 x + tanh2 x = 1, coth2 x = 1 + cosech2 x

23. sinh–1 x = log (x + x2 + 1 ), cosh–1 x = log ( x + x2 − 1 )

FG a − b IJ = tan a – tan b, tan F a + b I = tan

(1 – x)–1 = 1 + x + x2 + x3 + ... ; | x | < 1 (1 + x)–1 = 1 – x + x2 – x3 + ...

C+D C−D C+D C−D

C+D C−D C+D D−C

41. (sinh–1 x) = , (cosh–1 x) =

44. θ 0° 30° 45° 60° 90° 180° 270° 360°

z sin x dx = – cos x + c; z cos x dx = sin x + c

z tan x dx = log sec x + c; z cot x dx = log sin x + c

sec x dx = log (sec x + tan x) + c = log tan

sec x tan x dx = sec x + c; cosec x cot x dx = – cosec x + c

51. z sech2 x dx = tanh x + c, z cosech2 x dx = – coth x + c

z sinh x dx = cosh x + c, z cosh x dx = sinh x + c

z sech x tanh x dx = – sech x + c, z cosech x coth x dx = – cosech x + c

54. Leibnitz rule for differentiation under the integral sign

55. If r = xi + yj + zk then | r | = 2

71. Walli’s formula

72. (a sin bx – b cos bx) + c

73. Γ(1/ 2) = π , Γ(− 1/ 2) = – 2 π

75. sin nπ = 0 ; cos nπ = (– 1)n, sin n +

76. x3 + y3 + z3 – 3xyz = (x + y + z)(x2 + y2 + z2 – xy – yz – zx).

polar coordinates of P, then r =

number z and is denoted by z .

Let S be a non-empty set of complex numbers and δ be a positive real number.

2 A TEXTBOOK OF ENGINEERING MATHEMATICS