M6 2020 Normal Distribution Lecture Notes

CJC MATHEMATICS DEPARTMENT
2020 JC2 H2 MATHEMATICS (9758)

TOPIC: NORMAL DISTRIBUTION
H2 MATHEMATICS
TOPIC NORMAL DISTRIBUTION 2020/JC2
In the previous chapters, we have seen statistical experiments whose

outcomes take on discrete values, either finitely many or countably finitely
many. However, there are experiments whose outcomes may be neither
finite or countable. For example, when one conducts an investigation
measuring the height of a student. Assuming height to be a variable
measured to any degree of accuracy, then clearly we have an infinite number
of possible heights in the sample space that cannot be equated to the number
of whole numbers.
In this chapter, we shall direct our attention to continuous random variables
and explore a particular one that follows a special distribution, called the
Normal Distribution.
Source: https://goodgame43.wordpress.com/tag/height/
Mastery Learning Objectives: At the end of the chapter, students should be able to
 understand the concept of a normal distribution as an example of a continuous probability

normal model and its mean and variance; use of N   ,  2  as a probability model.
 make computations involving the standard normal distribution
 find the value of P  X  x1  or a related probability, given the values of x1 ,  ,  .
 make use of the symmetry of the normal curve and its properties to compute probabilities.
 determine a relationship between x1 ,  ,  given the value of P  X  x1  or a related
probability.
 solve problems involving the use of E  aX  b  and Var  aX  b  .
 solve problems involving the use of E  aX  bY  and Var  aX  bY  , where X and Y are
independent.
Reference:
Walpole, R.E et al. (2012). Probability & Statistics for Engineers & Scientists. (9th ed.).: Prentice Hall.
Pre-requisites:
□ Probability
□ Discrete Random Variables
1 Introduction to Continuous Random Variable
We first note the sample space for a continuous random variable.
If a sample space contains an infinite number of possibilities equal to the number of points on a line
segment, it is called a continuous sample space.
A random variable is called a continuous random variable when it can take on values on a
continuous scale.
Page | 146
For example, let X denote the random variable defined by proportion of residents in a town that vote
for a political party A. The random variable X takes on all values of x for which 0  x  1 .
In another example, let X denote the random variable defined by the waiting time, in hours, between
successive fatal accidents on a particular stretch of road. The random variable X takes on all values of
x for which x  0 .
Often, the possible values of a continuous random variable are precisely the same values that are
contained in the continuous sample space.
A continuous random variable, X , has a probability of P  X  x   0 . For example, let X denote the
random variable “height of a 21 year old man in Singapore”. For X to be a continuous random variable,
we need to make the assumption that we can measure height to any degree of accuracy. Hence, there
are infinitely many height measurements. As a result, a height of exactly 171 cm is just one of the
infinitely possible heights and P  X  171  0 . Consequently, the probability distribution of X cannot
be presented in a tabular form, unlike that of discrete random variables. It is usually stated as a function
in x, f  x  . Then the probability associated with X is given by the area under the graph of y  f  x  .
The function f  x is a probability density function for the continuous random variable X , defined
over the set of real numbers, if,
 f  x  0 for all x  ,

  f  x  dx  1 ,

b
 P a  X  b   f  x  dx .
a
__________________________________________________________________________________
 Remarks:
 The domain of f is usually taken to be  , with the understanding that f  x may be zero for
some x .
 P X  x  0 .
 P  a  X  b   P  a  X  b   P  a  X  b   P  a  X  b  , i.e. it does not matter whether
we include or exclude end points of the interval.
__________________________________________________________________________________
Page | 147
2 Introduction to Normal Distributions
The most important continuous probability distribution in the entire ﬁeld of statistics is the normal
distribution. Its graph, called the normal curve, is the bell-shaped curve as shown below, which
approximately describes many phenomena that occur in nature, industry, and research. For example,
physical measurements in areas such as meteorological experiments, rainfall studies and measurements
of manufactured parts are often more than adequately explained with a normal distribution. In addition,
errors in scientiﬁc measurements are extremely well approximated by a normal distribution.
In 1733, Abraham DeMoivre developed the mathematical equation of the normal curve. It provided a
basis from which much of the theory of inductive statistics is founded. The normal distribution is often
referred to as the Gaussian distribution, in honour of Karl Friedrich Gauss (1777–1855), who also
derived its equation from a study of errors in repeated measurements of the same quantity.
A continuous random variable X having the bell-shaped distribution of figure below is called a normal
random variable. The mathematical equation for the probability distribution of the normal variable
depends on the two parameters  and  , its mean and standard deviation, respectively.
Hence, we describe the distribution of X by writing X ~ N   , 2  .
[Adapted from Walpole, R.E et al. (2012). Pages 172-173]
A bell-shaped curve
The probability density function of a normally distributed random variable X with mean  and
variance  2 is
1
 2  x 
2
1
f  x  e 2 for   x  
2π
Properties of a Normal Curve

 The maximum value of the curve is at x   , therefore the mode is at  . Find f '  x  to verify.
 The curve is symmetrical about the line x   . As a result, the mean and median are both  .
 The curve has its points of inflexion at x     .
 It is concave downwards if     x     and is concave upward otherwise i.e. bell-shaped.
1
  x   2
 As x   , f  x   0 . You can observe this to be true due to the e 2 2
component of f  x  .
 1
  2  x
2
1
 Total area under the graph is 1. Note that  e 2 dx cannot be evaluated algebraically
 2π
easily.
Page | 148
The values of mean  and variance  2 affects the shape of the normal curve.
The mean  affects the location of the normal curve.
0 1 -2
X ~ N  0,1 X ~ N 1,1 X ~ N  2,1
The variance  2 affects the steepness/spread of the normal curve.
X ~ N  0,1 X ~ N  0, 0.5 X ~ N  0, 2 
Generally,
Approximately 68.3% of the Approximately 95.4% of the Approximately 99.7% of the

distribution lies within 1 distribution lies within 2 distribution lies within 3
standard deviation of the standard deviations of the standard deviations of the
mean. mean. mean.
Visit http://www.intmath.com/counting-probability/normal-distribution-graph-interactive.php to
explore the effect of  and  on the shape of the normal curve.
2.1 Finding Probabilities using the Graphing Calculator (G.C.)

Given that X ~ N  , 2 ,
b 1
  2  x 
2
1
P  a  X  b   e 2 dx .
a 2π
However, it is not easy to evaluate this algebraically. We can evaluate using the G.C.
Page | 149
Worked Example 1: If X ~ N  50,8 , find

(a) P  48  X  54 , (b) P  X  55 , (c) P  X  52  , (d) P  X  51  3
Solution:
(a) Press Screen Display Remarks
1. Press MODE and turn on STAT This is to activate
WIZARDS. wizard menu for
keying statistics
command.
2. Press 2ND VARS.
3. Select 2:normalcdf.
4. Input the following: To paste the G.C.

lower limit, upper limit,  and  . command
Then highlight Paste and press “normalcdf” with its
ENTER. inputs onto home
screen.
5. Press ENTER. To compute the value

of the probability.
 P  48  X  54  __________
0.682  3 s.f.
Page | 150
(b)
 P  X  55  __________
0.0385  3 s.f.
(c)
 P  X  52   __________
0.760  3 s.f.
(d) P  X  51  3  P  3  X  51  3
 P  48  X  54 
0.682
 __________  3 s.f.
__________________________________________________________________________________
 Remarks:
 To calculate P  X  a  , we set the lower limit as 1099 to represent negative infinity in the
G.C. Recall that the normal curve has a horizontal asymptote at y = 0.
 To calculate P  X  a  , we set the upper limit as 1099 to represent positive infinity in the G.C.
 The G.C. only recognises  for the input, and not  2 .
__________________________________________________________________________________
 Self-Practice:
If X ~ N  2, 3 , find (a) P 1.8  X  2.9  (b) P  X  2.8  
(c) P X  2  2 3 
Answers:
(a) 0.244 (b) 0.678 (c) 0.954
Page | 151
2.2 Finding Upper Limit Given Probability Using G.C.

Similarly, G.C. can be used to find the value of a given the value of P  X  a  , i.e. the area
under the graph which is shaded to the left of a .
P(𝑋 < 𝑎)
𝑎
Worked Example 2: If X ~ N 100,6  2
,
(a) find a given P  X  a   0.8907 .
(b) find b given P  X  b   0.770 .
Solution:
0.8907
(a) P  X  a   0.8907 can be illustrated on the normal curve.
a
Press Screen Display Remarks
1. Press 2ND VARS.
2. Select invNorm.
3. Input the GC OS 5.2.2 GC OS 5.1.0 To paste the G.C.

following: command “invNorm”
area,  and  . with its inputs onto
Then highlight home screen.
Paste and press
ENTER.
4. Press ENTER. GC OS 5.2.2 GC OS 5.1.0 To compute the value of

the limit.
 a  107  3 s.f . Page | 152

(b) P  X  b   0.770 0.770
b
GC OS 5.2.2 GC OS 5.1.0
b  95.6  3 s.f.  P  X  b   0.770
1  P  X  b   0.770
P  X  b   0.230
b  95.6  3 s.f. 
__________________________________________________________________________________
 Remarks:
When you are using GC OS 5.2.2, you can specify whether the area is to the left, centre or right.
When you are using GC OS 5.1.0,

 If you are given the area on the right side, i.e. P  X  b  , you will need to get the area on the
left side before using the G.C.
 Since P  X  b   0 , P  X  b   P  X  b  , then P  X  b   1  P  X  b   1  P  X  b  .
__________________________________________________________________________________
 Self-Practice:
 
If X ~ N 60, 52 , find
(a) m given P  X  m   0.2324 (b) n given P  X  n   0.837

Answers:
(a) 56.3 (b) 55.1
Page | 153
Worked Example 3: If X ~ N  70,25 , find the value of a such that P  X  70  a   0.8 .

Hence, find the limits within which the central 80% of the distribution lies.
Solution:
Using GC OS 5.2.2
P | X  70 | a   0.8
P   a  X  70  a   0.8 1 − 0.8
0.8
2
P  70  a  X  70  a   0.8
Using
70  a  63.592
a  6.4078
a  6.41  3 s.f.
Or 70
70  a  76.408
a  70  76.408
a  6.41  3 s.f.
Hence, the central 80% of the distribution lies between 63.6 and 76.4 (to 3 s.f.).
Using GC OS 5.1.0
P | X  70 | a   0.8
P   a  X  70  a   0.8
P  70  a  X  70  a   0.8
Using
P  X  70  a   0.1
70  a  63.592
a  6.4078
a  6.41  3 s.f.
Since a  6.4078
Left limit  70  a  63.6  3 s.f. 
Right limit  70  a  76.4  3 s.f. 
Hence, the central 80% of the distribution lies between 63.6 and 76.4.
You can check your answer by finding if P  63.6  X  76.4  is equals to 0.8.
 Remarks:
 It is usually a good idea to have a rough sketch of the area given before proceeding to solve.
 Many questions require you to use the symmetrical properties of a normal curve to solve.
__________________________________________________________________________________
Page | 154
3 Standard Normal Distribution
3.1 Definition
A normal distribution with mean 0 and variance 1 is called the standard normal distribution
and is denoted by Z ~ N  0,1 .
The probability density function of Z is

1  12 z 2
f  z  e for   z   .
2π
Worked Example 4: If Z ~ N  0,1 , find (i) P  0.32  Z  1.3

(ii) the value of a if P  Z  a   0.156 .
Solution:
(i)
P  0.32  Z  1.3  0.278  3s.f 
(ii) GC OS 5.2.2 GC OS 5.1.0

P  Z  a   0.156
a  1.01  3s.f 
 Self-Practice:
If Z ~ N  0, 1 , find
(a) P  2.56  Z  0.134  (b) P  Z  1.863 (c) P  Z  0.754 
Answers:
(a) 0.548 (b) 0.0312 (c) 0.451
Page | 155
3.2 Standardizing to a Standard Normal Distribution
In the past without a G.C., tables were used to evaluate probabilities involving the normal
variable. However, we cannot use a single table for all possible values of  and  . Thus, there
is a need to have the normal variable X standardized to a standard normal variable with a mean
of 0 and a standard deviation of 1 . This standard normal variable is denoted by Z . Though we
can use a G.C., the technique of standardizing a variable by hand is still useful in many situations,
such as finding unknown values of  or  .
To illustrate how a normal variable X is standardized to the standard normal variable Z , we

consider a random variable X , which is normally distributed with mean 5 and standard
1
  x 5 2
  1
e 2 2 
2
2
deviation 2 . i.e. X ~ N 5, 2 . The equation of this normal curve is y  .
2π  2 
1 ( )
( )
𝑦= e
√2π(2)
In order for the normal curve to have mean 0, we translate parallel to the x-axis 5 units in the
negative x-direction. The transformed function is obtained by replacing x with x  5 ,
1 2
  x  5   5 
1 2
e 
2 2
y
2π  2
1
  x 2
1 2 2 
2
y e
2π  2 
Page | 156
1 ( )
( )
𝑦= e
√2π(2)
1 ( )
( )
𝑦= e
√2π(2)
Now we scale the curve so that the standard deviation is 1 . This is done by:
1
 Scaling the graph parallel to the x-axis by a scale factor of . Replacing x with 2 x ,
2
1
  2 x 2
1 2 2 
2
y e
2π  2 
1
1   x 2
2(1) 2
y e
2π  2 
y
 Scaling the graph parallel to the y-axis by a scale factor of 2. Replacing y with ,
2
1
2 
2
 x
y 1
 e 21
2 2π  2 
1
  x 2
1 21
2
y e
2π 1
1 ( )
( )
1 ( ) 𝑦= e
𝑦= e ( ) √2π(2)
√2π(1)
1 ( )
( )
𝑦= e
√2π(2)
Page | 157
1
  x
2
1
f  x  e 21
2
is the probability density function of the standard normal with mean 0

2π 1
and variance 1.
Generally, given a normally distributed random variable X with mean,  , and variance,  2 ,
we can always transform it to the standard normal curve.
 
Given X ~ N  , 2 , its probability density function is represented by the curve
1
  x  2
1
e 2  
2
y
2π  
(1) Translate parallel to the x-axis   units. Replacing x with x   ,

1
  x     2
1 2 
2
y e
2 π  
1
  x 2
1 2 
2
y e
2 π  
1
(2) Scale parallel to the x-axis by a scale factor of . Replacing x with  x ,

1
  x 2
1 2  
2
y e
2 π  
1
  x 2
1 21
2
y e
2 π  
y
(3) Scale parallel to the y-axis by a scale factor of  . Replacing y with ,

1
  x 2
y 1 21
2
 e
 2π  
1
  x 2
1 21
2
y e
2π 1
 
If X ~ N  , 2 and Z ~ N  0,1 then
x  x  X 
P  x1  X  x2   P  1 Z 2 and Z .
    
Page | 158
Graphical Reasoning
The figure below show the area of P  x1  X  x2  . Let this area be A .
𝐴
𝑂 𝜇 𝑥 𝑥
When we transform X to Z , we need to do the following three transformations in order.
(1) Translate parallel to the x-axis   units. The area remains as A .
𝑥 −𝜇 𝑥 −𝜇
𝐴
𝑂
1 A
(2) Scale parallel to the x-axis by a scale factor of . The area is now .
 
𝜎 𝜎
𝐴
𝜎
𝑂
Page | 159
A
(3) Scale parallel to the y-axis by a scale factor of  . The area is now   A .

𝜎 𝜎
x  x 
Therefore, P  x1  X  x2   P  1 Z  2 .
   
Since
 x  x 
P  x1  X  x2   P  1 Z  2
   
x  x 
P  x1    X    x2     P  1 Z  2
   
 x   X   x2     x  x 
P 1     P 1 Z  2 .
        
X 
Comparing, Z  .

 
Worked Example 5: If X ~ N  , 62 and P  X  82   0.0478 , find the value of the mean  .
Solution:
Given P  X  82   0.0478 0.0478
GC OS 5.2.2
GC OS 5.1.0
P  X  82   0.0478
P  X  82   0.0478
 82   
P Z    0.0478 P  X  82   1  0.0478
 6 
82    82   
 1.66657 P Z    0.9522
6  6 
  72.0  3 s.f. 82  
 1.66657
6
  72.0  3 s.f. 
Page | 160
Worked Example 6: The masses of Blue Neon Guppies bred by Green Park Tropical Fish Farm are
normally distributed with mean  g and standard deviation  g. 5% of the fish
have a mass greater than 85g and 10% have a mass less than 25g.
Find the values of  and  , and find the range symmetrical about the mean,
within which 75% of the masses lie.
Solution:
Let X be the random variable denoting the mass of a randomly selected Blue Neon Guppies in grams.

X ~ N  , 2 
P  X  25  0.1
 25    0.1
P Z   0.1
  
25  
 1.281551567

  1.281551567  25      1
P  X  85   0.05
GC OS 5.1.0
0.05 P  X  85  0.05
 85   
P Z   0.05 P  X  85  1  0.05
  
85    85   
P Z   0.95

 1.64485
  
  1.64485  85        (2) 85  
 1.64485

  1.64485  85        (2)
Solving equations  and ,   51.2756  51.3  3 s.f.

  20.5030  20.5  3 s.f.
0.75
0.125

X ~ N 51.2756, 20.50302 
Let P  a  X  b   0.75
Page | 161
GC OS 5.2.2
GC OS 5.1.0
1  0.75
P X  a   0.125
2
a  27.7  3 s.f. 
P  X  b   1  0.125
b  74.9  3 s.f. 
Hence, 75% of the masses lie between 27.7g and 74.9g.
__________________________________________________________________________________
 Remark: If  or  or both are unknown, we need to standardize X ~ N ,  2 to  
Z ~ N  0, 1 before we can infer from given probabilities.
__________________________________________________________________________________
4 Linear Transformation of a Normal Random Variable
 
If X ~ N  ,  2 , then, for any real constants a and b , aX  b and aX  b will follow a normal
distribution with
aX  b ~ N  a  b, a 2 2  .
5 Linear Combination of Independent Normal Variables
If X and Y are two independent normal distributions, such that X ~ N  X ,  X 2   and
 
Y ~ N Y ,  Y 2 , then for any real constants a and b , aX  bY and aX  bY will follow a normal
distribution with
aX  bY ~ N  a  X  bY , a 2 X 2  b2 Y 2  .
6 Sum of Independent and Identically Distributed Normal Variables
 
If X ~ N  ,  2 and that X 1 , X 2 , X 3 ,..., X n are n independent random observations of X , then
X 1  X 2  X 3  ...  X n follows a normal distribution with

X 1  X 2  X 3  ...  X n ~ N n , n 2 .
Page | 162
 Remark: It is important to know that nX  X 1  X 2  X 3    X n .
nX ~ N  n , n 2 2 
X1  X 2  X 3    X n ~ N  n , n 2 
__________________________________________________________________________________
Worked Example 7: X and Y are independent random variables such that X ~ N 100, 49  and
Y ~ N 110, 576  . Find
(i) P  3 X  250  (ii) P  X  Y  200 
(iii) P Y  X  (iv) P  2 X  Y  300 
Solution:
(i) Method : Method :

 250 

3 X ~ N 3  100,32  49  P  3 X  250   P  X 
 3 

3 X ~ N  300,441  0.00863  3 s.f.

P  3 X  250   0.00863  3 s.f. 
(ii) X  Y ~ N 100  110, 49  576 

X  Y ~ N  210,625
P  X  Y  200   0.655  3 s.f. 
(iii) Y  X ~ N 110  100,576  49 

Y  X ~ N 10,625
P Y  X  =P Y  X  0   0.655  3 s.f.
(iv)   
2 X  Y ~ N  2  100   110, 22  49  576 
2 X  Y ~ N  310,772 
P  2 X  Y  300   0.359  3 s.f. 
Page | 163
Worked Example 8: In a certain country, the weights of men are normally distributed with mean 70
kg and variance 18 kg2. The weights of children are normally distributed with
mean 30 kg and variance 9 kg2. Find the probability that
(i) the weight of a randomly selected man is more than the combined weights
of two randomly selected children,
(ii) the weight of a randomly selected man is more than twice the weight of a
randomly selected child.
Solution:
(i) Let M and C be the random variables denoting the weight of a randomly selected man and a
randomly selected child respectively, in kg.
M ~ N  70, 18  and C ~ N  30, 9 
M  C1  C 2 ~ N  70  30  30, 18  9  9 
M  C1  C 2 ~ N 10, 36 
P  M  C1  C 2   P  M  C1  C 2  0 
 0.952  3 s.f.
(ii) 
M  2C ~ N 70  2  30  ,18  22  9  
M  2C ~ N 10, 54 
P  M  2C   P  M  2C  0 
 0.913 (3s.f.)
__________________________________________________________________________________
 Remark: Confusion often arises over the random variables 2X and X 1  X 2 , where X 1 and X 2
are independent observations of the random variable X . Usually, words like “twice”,
“thrice” mean scaling while words like “sum” usually means summing up independent
observations.
__________________________________________________________________________________
 Self-Practice:
If B and G are random variables denoting the height of a randomly selected boy and a randomly
selected girl respectively, use a combination of B and G to describe the following:
(a) total height of two boys and two girls,
(b) total height of two boys and twice the height of one girl,
(c) total height of 10 girls and thrice the height of two boys.
Assume all boys and girls are selected randomly.
Answers:
(a) B1 + B2 + G1 + G2 (b) B1 + B2 + 2G (c) G1 + … + G10 + 2(B1 + B2)
Page | 164
Worked Example 9: The maximum load a certain HDB lift can carry is 340 kg. The weights of men
are normally distributed with mean 60 kg and standard deviation 10 kg. The
weights of women are normally distributed with mean 55 kg and standard
deviation 5 kg. Find the probability that the lift will be overloaded by 3 men and
2 women if their weights are independent.
Solution:
Let M and W be the random variables denoting the weight of a randomly selected man and a randomly
selected woman respectively, in kg.

M ~ N 60, 102  and 
W ~ N 55, 52  P(lift will be overloaded by 3 men and 2 women)
= P(M1 + M2 + M3 + W1+ W2 > 340)
M ~ N  60, 100  W ~ N  55, 25
M 1  M 2  M 3  W1  W2 ~ N  3  60  2  55, 3  100  2  25 
M 1  M 2  M 3  W1  W2 ~ N  290, 350 
P  M 1  M 2  M 3  W1  W2  340   0.00376  3s.f.
Worked Example 10: In a certain nation, men have heights distributed normally with mean 1.70 m
and standard deviation 10 cm; women have heights distributed normally with
mean 1.60m and standard deviation 7.5 cm.
(i) Find the probability that a man chosen at random has height not less than 1.83m.
(ii) Two men are selected at random. Find the probability that both of them will be taller than 1.83 m.
(iii) Five men are selected at random. Find the probability that exactly 3 of them will be taller than 1.83 m.
(iv) What is the probability that the average height of 3 men chosen randomly is greater than 1.78 m?
(v) Find the probability that a husband and wife have not more than 5 cm difference in heights. State
the assumptions that you have made in the calculation.
Solution:
(i) Let M and W be the random variables denoting the heights of a man and a woman
respectively, in cm. M ~ N 170, 10 2 and  
W ~ N 160, 7.52  
P  M  183   0.0968005  0.0968  3 s.f.
(ii) P  M 1  183  P  M 2  183  0.09680052 =0.00937  3 s.f.
(iii) Let X be the random variable denoting the number of men who will be taller than 1.83 m out
of 5 men. X ~ B  5,0.0968005494 
P  X  3  0.00740  3 s.f.
(iv) M 1  M 2  M 3 ~ N 170  170  170, 100  100  100 

M 1  M 2  M 3 ~ N  510,300 
 M  M2  M3 
P 1  178   P  M 1  M 2  M 3  534 
 3 
 0.0829  3 s.f.
Page | 165
(v) 
M  W ~ N 170  160,102  7.52 
M  W ~ N 10,156.25
P  | M  W | 5   P  5  M  W  5 
 0.230  3 s.f.
We assume that the height of a randomly selected husband and a randomly selected wife are
independent.
7 Learning Experiences (Not in syllabus)
Normal Approximation to Binomial Distribution
Under certain conditions, a normal distribution can be used to approximate a binomial distribution,
X ~ B  n, p  . The diagram below showed the approximation when we keep np constant and increase
the values of n .
np  2
n  10 n  50 n  100
np  5
n  10 n  50 n  100
Generally, a normal distribution is a good approximate to a binomial distribution when np  5

and n 1  p   5 .
You can explore the relationship for different values of n and constant np at
http://www.byebyemaths.net/statistics/binomial-to-normal.
Page | 166
SUMMARY Normal Distribution
Fill in the following blanks to recap what you have learnt in this topic.
Let X be the random variable denoting “________ of a _______”

Notation
X ~ N   , 2 
Expectation,
Variance, EX    Var  X    2 Standard deviation  
Standard Deviation

 
Properties (1) The normal distribution curve is bell shaped and symmetrical about the line
x.
(2) The total area under a normal curve is 1.
(3) The area under a normal curve between x  a and x  b gives the value of the
probability P  a  X  b  .
(4) The shape and location of a normal curve are determined by the values of 
and  .
If   0 . and   1 , then Z ~ N  0, 1 is called the standard normal

distribution.
Standardization
If  and  or both are unknown, there is a need to standardize X ~ N  ,  2  
before the probabilities can be computed.
X 

If X ~ N  ,  2  and Z 

, then Z ~ N  0, 1
If X and Y are two independent normal variables such that

  
X ~ N  X ,  X 2 and Y ~ N Y ,  Y 2 , then 

X  Y ~ N  X  Y ,  X 2   Y 2 
Sum of Independent X Y ~ N X  Y ,  X   Y
2 2

Normal Variables

aX  bY ~ N a  X  bY , a 2 X 2  b 2 Y 2 

aX  bY ~ N a  X  bY , a 2 X 2  b 2 Y 2 
Page | 167
If X ~ N   ,  2  and that X 1 , X 2 , X 3 ,..., X n are n independent random

observations of X , then X 1  X 2  X 3  ...  X n follows a normal distribution with
Sum of Independent
and Identically

X 1  X 2  X 3  ...  X n ~ N n , n 2 
Distributed Normal
Variables 
If X ~ N  ,  2  and given that nX is n times of one random observation of
X , then nX follows a normal distribution with

nX ~ N n , n 2 2 
Page | 168
H2 MATHEMATICS TUTORIAL
TOPIC NORMAL DISTRIBUTION 2020/JC2
DISCUSSION
BASIC LEVEL
1. Packages from a packing machine have a mass which is normally distributed with mean 200g
and standard deviation 2g.
Find the probability that a package from the machine weighs
(a) between 198.5 and 199.5g
(b) less than 197g
(c) more than 200.5g
2. [2008/H1/7(a)]
An examination is marked out of 100. It is taken by a large number of candidates. The mean
mark, for all candidates, is 72.1, and the standard deviation is 15.2. Give a reason why a normal
distribution, with this mean and standard deviation, would not give a good approximation to
the distribution of marks.
3. The heights of boys and girls in Catholic Junior College may be assumed to follow independent
normal distributions. For each of the following cases, state with a reason, whether or not a normal
model is likely to be appropriate.
(a) The height of a student chosen at random from a combined group of CJC boys and girls.
(b) The sum of heights of a randomly chosen sample of 3 CJC boys.
(c) The number of days since his last birthday of a randomly chosen CJC student.
4. If X ~ N  75, 5  , Y ~ N  78, 20  and X and Y are independent, find

(i) P  X  Y  155  ,
(ii) P Y  X  15  .
5. [N87/II/9]
X and Y are continuous random variables having independent normal distributions.
The means of X and Y are 10 and 12 respectively, and the standard deviations are 2 and 3
respectively.
Find (a) P  Y  10  (b) P  2 X  25 

(c) P Y  X  (d) P  4 X  5Y  90 
(e) the value of x such that P  X 1  X 2  x   0.25 , where X 1 and X 2 are
independent observations of X .
Page | 169
INTERMEDIATE LEVEL
6. [2007/HCI/I/6]
A manufacturer produces packets of margarine. The mass of margarine in a packet has a normal
distribution of mean 502 g and standard deviation 0.8 g. Find the proportion of packets which
contain less than 500 g of margarine.
The manufacturer increases the mean amount of margarine in a packet to  g. The standard
deviation remains unchanged. Only 1 packet in 1000, on average, now contains less than 500 g.
Find,  , correct to 1 decimal place.
7. [SP2006/II/9(modified)]
A grading machine is set to reject an apple as too large if its mass exceeds 150 g, and too small
if its mass is less than 110 g. Over a long period, 10% of apples are rejected as too large and
15% are rejected as too small. The mass, in grams, of a randomly chosen apple may be taken to
 
have the distribution N  ,  2 . Find  and  .
8. [2009/MI/BT/II/9(a)(modified)]
The random variable X has the distribution N(  , 2 ) . Given that
P( X  85)  P( X  101)  0.0548 , find the values of  and  .
9. Lengths of rods of type A are normally distributed with mean 5 cm and standard deviation
0.5 cm and lengths of rods of type B are normally distributed with mean 10 cm and standard
deviation 1 cm. Assuming the lengths of rods of type A and B are independent. Find the
probability that
(a) a length consisting of 3 rods of type A and 2 rods of type B is between 33 cm and
36 cm long,
(b) a length consisting of 6 rods of type A is longer than three times the length of a
randomly chosen rod of type B .
10. [J87/II/8]
The weight of the contents of a randomly chosen packet of breakfast cereals A may be taken to
have a normal distribution with mean 625 g and standard deviation 15 g. The weight of the
packaging may be taken to have an independent normal distribution with mean 25 g and standard
deviation of 3 g. Find, giving to three significant figures in your answers,
(a) the probability that a randomly chosen packet of A has a total weight exceeding 630 g,
(b) the probability that the total weight of the contents of 4 randomly chosen packets of A
exceeds 2450 g.
The weight of the contents of a randomly chosen packet of breakfast cereals B may be taken to
have a normal distribution with mean 465 g and standard deviation 10 g. Find the probability that
the contents of four randomly chosen packets of B weigh more than the contents of three randomly
chosen packets of A .
Page | 170
ADVANCED LEVEL
11. A variable X is distributed normally with mean 10 and standard deviation 2 .

Find P  9  X  13 and P  X  12  .
Find also the possible values of a if the probability that X is between a and 12 is 0.15.
If three values of X are taken at random, calculate the probability that two of them are less
than 12 and the other is greater than 12.
12. [N07/II/8]
Chickens and turkeys are sold by weight. The masses, in kg, of chickens and turkeys are modeled as
having independent normal distributions with means and standard deviations as shown by the table.
Mean Mass Standard Deviation

Chickens 2.2 0.5
Turkeys 10.5 2.1
Chickens are sold at $3 per kg and turkeys at $5 per kg.

(i) Find the probability that a randomly chosen chicken has a selling price exceeding $7.
(ii) Find the probability of the event that both a randomly chosen chicken has a selling price
exceeding $7 and a randomly chosen turkey has a selling price exceeding $55.
(iii) Find the probability that the total selling price of a randomly chosen chicken and a
randomly chosen turkey is more than $62.
(iv) Explain why the answer to part (iii) is greater than the answer in part (ii).
13. [2008/VJC/BT/II/14]
  
Two independent random variables X and Y have the distributions N 2 ,  2 and N 3 , 3 2 
respectively.
(i)  
Given that P X  2  2 2  k , show that P  Z  2   k .
(ii)  
Write down the mean and the variance of Y  X . Hence find P Y  X    4 2 in terms
of k.
14. [N89/II/9]
The random variable X has a normal distribution with mean 3 and variance 4. The random
variable S is the sum of 100 independent observations of X , and the random variable T is the
sum of a further 300 independent observations of X . Giving your answers to 3 decimal places,
(i) P  S  310  ,
(ii) P  3S  50  T  ,
(iii) The random variable N is the sum of n independent observations of X . State the
approximate value of P  N  3.5n  as n becomes very large, justifying your answer.
Page | 171
Answers:
1. (a) 0.175 (b) 0.0668 (c) 0.401
4. (i) 0.655 (ii) 0.992
5. (a) 0.252 (b) 0.106 (c) 0.290 (d) 0.722 (e) 21.9
6. 0.00621; 502.5
7. 128; 17.3
8.   93 ,   5.00
1
9. (a) 0.613 (b) X ~ N(0,10.5) ;
2
10. (a) 0.904 (b) 0.952; 0.324
11. 0.625; 0.841; 11.0 or 14.8; 0.337
12. (i) 0.395 (ii) 0.160 (iii) 0.392
13. 2 1  k 
14. (i) 0.309 (ii) 0.235 (iii) 0
REVIEW PROBLEMS
1. [2009/TPJC/II/11]
The times taken by two runners A and B to run 400 metre races are independent and normally
distributed with means 45.0 seconds and 45.2 seconds and standard deviation 0.5 seconds and
0.8 seconds respectively. The two runners are to compete in a 400 metre race for which there is
a track record of 44.5 seconds.
(i) Calculate the probability of runner A breaking the track record. Hence, show that the
probability of runner B breaking the record is greater than that of runner A. [3]
(ii) Find the probability of runner A beating runner B. [3]
(iii) Find the probability that the total sum of four randomly chosen times of A is more than 4
times a randomly chosen time of B. [5]
2. [2009/IJC/II/9]
Durians are sold at $7 per kilogram. The mass of a randomly chosen durian is normally
distributed with mean 1.5 kg and standard deviation 0.2 kg. Mangoes, which are sold in bags of
five, are priced at $2 per kilogram. The mass of a randomly chosen mango is normally distributed
with mean 0.4 kg and standard deviation 0.05 kg. The masses of durians and mangoes may be
assumed to be independent of each other.
(i) Find the probability that a randomly chosen bag of mangoes has a selling price exceeding
$4.20. [3]
(ii) Find the probability of the event that both a randomly chosen durian has a selling price
exceeding $10 and a randomly chosen bag of mangoes has a selling price exceeding $4.20.
[3]
(iii) Find the probability that the total selling price of 2 durians and 3 bags of mangoes is less
than $33.50. [5]
[The mass of an empty bag is assumed to be negligible.]
Page | 172
3. [2009/TJC/II/11]
A factory produces a particular brand of oil in bottles of 3 sizes namely Small, Regular and Large.
The amount of oil in each type of bottle, in millilitres, is assumed to be normally distributed as
shown in the table below.
Mean (ml) Variance (ml2)
Small 610 25
Regular 1050 2
Large 1200 50
(i) Explain why the amount of oil in a Small bottle can be modelled by a normal distribution
even though amount of oil cannot possibly be negative. [1]
(ii) Three Small bottles are randomly chosen. Find the probability that one of the Small bottles
contains more than 615 ml of oil while the other two Small bottles contain less than 602
ml of oil. [3]
(iii) Find the value of  such that 80% of Regular bottles contain between 1040 ml and 1060
ml of oil. [3]
(iv) Find the probability that the total amount of oil contain in 2 randomly chosen Large bottles
exceeds 4 times the amount of oil contain in a randomly chosen Small bottle. [3]
4. [2009/AJC/II/13(part of)]
A telco company found that the duration of calls made by its customers to City A are normally
distributed with mean 8 minutes and standard deviation 2.5 minutes. The duration of calls made
by its customers to City B are normally distributed with mean 10 minutes and standard deviation
3.2 minutes. The duration of any call is independent of the duration of any other calls.
(i) Find the probability that the total duration of two randomly selected calls to City A differs
from thrice the duration of a third call to City A by more than three minutes. [3]
The company charges a rate of 22 cents per minute for every call to City A and a rate of 30 cents
per minute for every call to City B. For each call, customers are also charged a handling fee of
10 cents. Ann makes two calls to City A and one call to City B.
(ii) Find the total amount of money she is expected to pay for the calls. [1]
(iii) Show that the probability that she spends more than $8.50 on the calls is 0.0870, corrected
to 3 significant figures [2]
5. [2009/NJC/II/12(part of)]
On Crabby Island, a fishmonger rears many types of live seafood in his kelong, one of which is
a particular type of crab named the Hairy Crab. The weight H, in kg, of a random Hairy Crab
follows a normal distribution with mean  kg and standard deviation  kg.
Given that P( H  1.5)  P( H  3.5)  0.33 , state the value of  . [1]
(i) Show that   2.27 . [2]
(ii) Find the probability that out of four randomly chosen crabs, two weigh between 1.5 kg and
3.5 kg each and the other two weigh more than 3.5 kg each. [3]
Another type of crab named Flower Crab found on Crabby Island follows an independent normal
distribution with mean weight 3.2 kg and standard deviation 1.3 kg. Find the largest value of a
such that the probability that the average weight of two randomly chosen Hairy Crabs and three
randomly chosen Flower Crabs exceeds a kg is more than 0.6. [3]
Page | 173
6. [2009/ACJC/II/11(part of)]
Peter Piper the Pickle Packer provides a written guarantee with each bottle of his pickles that
every pickle will be at least 8 cm long. His bottles are made such that the longest pickle that they
can hold is 10.5 cm. The lengths of the pickles are normally distributed with mean of 9 cm and
standard deviation 0.8 cm.
(i) Find the probability that a randomly chosen pickle Peter Piper picked will have to be
discarded because
(a) it does not meet his guaranteed standard; [1]
(b) it does not fit the bottle. [1]
(ii) Peter wishes to discard no more than 5 percent of the pickles he picks. Comment on the
validity of this statement [1]
(iii) Find the probability that if Peter randomly picks two pickles, both pickles will fail to meet
his guaranteed standard given that their total length is less than 16 cm. [3]
7. [2009/NYJC/II/6(modified)]
The height of girls, chosen at random, is found to be normally distributed with mean of 155cm
with a standard deviation of 10cm. Find
(i) the range of the girls’ height if 40% of the girls have heights falling symmetrically within
the mean, [3]
(ii) the probability that exactly two girls are taller than 170cm if four girls are chosen, [3]
(iii) the probability that the total height of n randomly chosen girls is more than 150n cm, if n
is large. [3]
8. [2009/RI/II/8]
The random variable X is normally distributed with mean 20 and variance  2 .
(i) Given that P(12  X  k )  P( X  k )  0.42065 , show that  = 8, correct to the nearest
whole number, and find the value of the constant k. [4]
X is related to a normal random variable Y by the formula X  Y1  Y2  4 , where Y1 and Y2 are
two independent observations of Y. Another random variable W, where W and Y are independent,
is normally distributed with mean 25 and variance 16.
(ii) Using  = 8, find E(Y ) and Var(Y ) , and hence find P(W  2Y ) . [5]
9. [2017/HCI/II/5]
The independent random variables X and Y are normally distributed with the same mean
7 but different variances Var( X ) and Var(Y ) , respectively. It is given that
P( X  10)  P(Y  6) .
(i) Show that Var  X   9Var Y  . [3]
(ii) If Var Y   1 , find P( X  9) . [2]
Page | 174
10. [2017/VJC/II/8]
In this question, you should state clearly the values of the parameters of any normal distribution
you use.
A supermarket sells two types of durians, D25 and Musang Queen. The durians are sold by
weight. The masses, in kilograms, of D25 and Musang Queen are modelled as having normal
distributions. The means and standard deviations of these distributions, and the selling prices, in
$ per kilogram, are shown in the following table.
Mean Standard deviation Selling price

(kg) (kg) ($ per kg)
D25 1.5 0.02 9
Musang Queen 1.8 0.035 18
(i) A customer buys 3 D25 durians and 2 Musang Queen durians. Find the probability that
the total cost of his purchase is more than $107. [5]
(i) State an assumption needed for your calculations in part (i). [1]
(ii) The probability that the average weight of n randomly chosen D25 durians exceeding m
kg is at least 0.1. Show that n satisfies the inequality
 m 1.5 n  0.025631 .
Hence find the largest possible value of n when m  1.51 . [4]
11. [2018/EJC/MYE/11]
In this question, you should state clearly the values of the parameters of any normal distribution
you use.
A supermarket sells grapefruits and papayas by mass. The masses, in grams, of a grapefruit and
a papaya are modelled as having normal distributions. The means and standard deviations of
these distributions, and the selling prices, in $ per gram, are shown in the following table.
Mean mass Standard deviation Selling price
(g) (g) ($ per g)
Grapefruit  15 0.007
Papaya 450 20 0.008
(i) Given that 65% of the grapefruits sold weigh more than 300 g, show that   306 , leaving
your answer to the nearest whole number. [2]
Take   306 for the rest of the question.
(ii) Find the probability that the total mass of 3 randomly chosen grapefruits is less than 950 g.
[2]
(iii) Ten grapefruits are chosen at random. Find the probability that more than half of the
grapefruits weigh more than 300g. [2]
(iv) Find the probability that the total mass of 3 randomly chosen grapefruits differs from the
total mass of 2 randomly chosen papayas by at most 100 grams. [3]
(v) Find the probability that it will cost at least $9 to buy 4 grapefruits. [3]
(vi) The supermarket decides to have a promotion, selling a bundle of 1 grapefruit and 1 papaya
at a fixed price. Find the least price, $a, that the supermarket should charge for the bundle
if there is at least a 90% chance of not making a loss. [3]
Page | 175
12. [2018/AJC/MYE Paper 2/6]

A confectionary bakes and sells butter and chocolates cookies. The masses, in grams, of butter
and chocolate cookies are modelled as having independent normal distributions with means and
standard deviations as shown in the table.
Mean mass Standard deviation
Butter cookies 6 0.2
Chocolate cookies 10 0.3
(i) What percentage of the butter cookies weighs more than 6.3g? [1]
(ii) Five chocolate cookies are randomly chosen. Find the probability that the lightest of the
five chocolate cookies is more than the mean mass of the butter cookies by more than 4g.
[3]
(iii) Butter cookies are sold in packets of 10 cookies each. The packaging is of negligible weight.
Find the probability that the mass of a randomly selected packet of 10 butter cookies weighs
more than 6 times the mass of a chocolate cookie by at least 3g. [4]
(iv) The confectionary also sells assorted packets of cookies containing a mixture of a total of
10 butter and chocolates cookies. If there are n chocolate cookies in an assorted packet,
find the least value of n such that there is a probability of at least 0.5 that the mass of an
assorted packet exceeds a packet of 10 butter cookies by more than 13g. [5]
Answers:
1. (i) 0.159 (ii) 0.584 (iii) 0.406
2. (i) 0.186 (ii) 0.119 (iii) 0.598
3. (i) P  S  0   0 (ii) 0.00143 (iii) 7.80 (iv) 0.0368
4. (i) 0.819 (ii) $6.82
5. 2.5 (ii) 0.0755; a  2.72
6. (i)(a) 0.106 (b) 0.0304 (iii) 0.290
7. (i) 5.24 (ii) 0.0233 (iii) 1
8. (i)   8  nearest integer  ; k  21.6  3s.f  (ii) 0.533
9. (ii) 0.748
10. (i) 0.0359 (iii) 6
11. (i)   305.78 (ii) 0.891 (iii) 0.751 (iv) 0.983 (v) 0.0198 (vi) least price is $5.99
12. (i) 6.68% (ii) 0.03125 (iii) 0.0579 (iv) 4
Page | 176

M6 2020 Normal Distribution Lecture Notes

Uploaded by

Copyright:

Available Formats

M6 2020 Normal Distribution Lecture Notes

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

M6 2020 Normal Distribution Lecture Notes

Uploaded by

Copyright:

Available Formats

CJC MATHEMATICS DEPARTMENT

2020 JC2 H2 MATHEMATICS (9758)

In the previous chapters, we have seen statistical experiments whose

 understand the concept of a normal distribution as an example of a continuous probability

1 Introduction to Continuous Random Variable

We first note the sample space for a continuous random variable.

2 Introduction to Normal Distributions

Properties of a Normal Curve

The mean  affects the location of the normal curve.

The variance  2 affects the steepness/spread of the normal curve.

Approximately 68.3% of the Approximately 95.4% of the Approximately 99.7% of the

2.1 Finding Probabilities using the Graphing Calculator (G.C.)

Worked Example 1: If X ~ N  50,8 , find

2. Press 2ND VARS.

4. Input the following: To paste the G.C.

5. Press ENTER. To compute the value

2.2 Finding Upper Limit Given Probability Using G.C.

3. Input the GC OS 5.2.2 GC OS 5.1.0 To paste the G.C.

4. Press ENTER. GC OS 5.2.2 GC OS 5.1.0 To compute the value of

 a  107  3 s.f . Page | 152

(b) P  X  b   0.770 0.770

When you are using GC OS 5.1.0,

(a) m given P  X  m   0.2324 (b) n given P  X  n   0.837

Worked Example 3: If X ~ N  70,25 , find the value of a such that P  X  70  a   0.8 .

3 Standard Normal Distribution

The probability density function of Z is

Worked Example 4: If Z ~ N  0,1 , find (i) P  0.32  Z  1.3

(ii) GC OS 5.2.2 GC OS 5.1.0

3.2 Standardizing to a Standard Normal Distribution

To illustrate how a normal variable X is standardized to the standard normal variable Z , we

is the probability density function of the standard normal with mean 0

(1) Translate parallel to the x-axis   units. Replacing x with x   ,

When we transform X to Z , we need to do the following three transformations in order.

(1) Translate parallel to the x-axis   units. The area remains as A .

Solving equations  and ,   51.2756  51.3  3 s.f.

4 Linear Transformation of a Normal Random Variable

5 Linear Combination of Independent Normal Variables

If X and Y are two independent normal distributions, such that X ~ N  X ,  X 2   and

6 Sum of Independent and Identically Distributed Normal Variables

 Remark: It is important to know that nX  X 1  X 2  X 3    X n .

(i) Method : Method :

3 X ~ N  300,441  0.00863  3 s.f.

(ii) X  Y ~ N 100  110, 49  576 

(iii) Y  X ~ N 110  100,576  49 

P  M 1  M 2  M 3  W1  W2  340   0.00376  3s.f.

(ii) P  M 1  183  P  M 2  183  0.09680052 =0.00937  3 s.f.

(iv) M 1  M 2  M 3 ~ N 170  170  170, 100  100  100 

7 Learning Experiences (Not in syllabus)

Normal Approximation to Binomial Distribution

Generally, a normal distribution is a good approximate to a binomial distribution when np  5

SUMMARY Normal Distribution

Let X be the random variable denoting “________ of a _______”

If   0 . and   1 , then Z ~ N  0, 1 is called the standard normal

If X and Y are two independent normal variables such that

If X ~ N   ,  2  and that X 1 , X 2 , X 3 ,..., X n are n independent random

4. If X ~ N  75, 5  , Y ~ N  78, 20  and X and Y are independent, find

Find (a) P  Y  10  (b) P  2 X  25 

11. A variable X is distributed normally with mean 10 and standard deviation 2 .

Mean Mass Standard Deviation

Let X be the random variable denoting “__ of a _”