Lecture 13

Joint Probability Distributions

Definition: The function f(x, y) is a joint density function of the continuous random variables X
and Y if

1. f(x, y) ≥ 0, for all (x, y),

∞ ∞

2. ∫∫
−∞ −∞
f ( x, y )dxdy = 1,

3. P[(X, Y) ∈ A] = ∫∫ f ( x, y )dxdy, for any region A in the xy plane.

A

Definition: The marginal distributions of X alone and of Y alone are

∞ ∞

=g ( x) ∫=
f ( x, y )dy and h( y ) ∫
−∞ −∞
f ( x, y )dx

for the continuous case.

Statistical Independence

Definition: Let X and Y be two random variables, discrete or continuous, with joint probability
distribution f(x, y) and marginal distributions g(x) and h(y), respectively.
The random variables X and Y are said to be statistically independent if and only if

f(x, y) = g(x)h(y)
for all (x, y) within their range.

Definition: Let X 1 ,X 2 , . . . , X n be n random variables, discrete or continuous, with joint

probability distribution f(x 1 , x 2 , . . . , x n ) and marginal distribution f 1 (x 1 ), f 2 (x 2 ), . . . ,
f n (x n ), respectively. The random variables X 1 ,X 2 , . . . , X n are said to be mutually
statistically independent if and only if
f(x 1 , x 2 , . . . , x n ) = f 1 (x 1 )f 2 (x 2 ) · · · f n (x n )
for all (x 1 , x 2 , . . . , x n ) within their range.

Example 12.2: A privately owned business operates both a drive-in facility and a walk-in
facility. On a randomly selected day, let X and Y, respectively, be the proportions
of the time that the drive-in and the walk-in facilities are in use, and suppose that
the joint density function of these random variables is
2
 (2 x + 3 y ), 0 ≤ x ≤ 1,0 ≤ y ≤ 1,
f(x, y) =  5
 0, elsewhere.
PHM111s - Probability and Statistics
 (a) Verify that f(x, y) is a joint density function.
1 1 1
(b) Find P[(X, Y) ∈ A], where A = {(x, y) | 0 < x < , < y < }.
2 4 2
(c) Find the marginal distribution of X alone and of Y alone g(x) and h(y).

Solution: (a) The integration of f(x, y) over the whole region is

∞ ∞ 1 1
2
∫−∞ −∞∫ f ( x, =
y )dxdy ∫ ∫ (2 x + 3 y )dxdy
0 0 5
1
2 x 2 6 xy x =1
= ∫( + ) |x = 0 dy
0 5 5
1
2 6y 2 y 3y2 1 2 3
= ∫( + )dy = ( + ) |0 = + = 1
0 5 5 5 5 5 5

(b) To calculate the probability, we use

1 1 1
P[(X, Y) ∈ A] = P (0 < X < , < Y < )
2 4 2
1/2 1/2
2
= ∫ ∫ (2 x + 3 y )dxdy
1/4 0 5

1/2
2 x 2 6 xy x =1/2 1/2
1 3y
∫1/4 5
= ( +
5
) | x=0
dy ∫
=+ (
1/4 10 5
)dy

y 3 y 2 1/2 13
=
( + ) |1/4 = .
10 10 160

(c) By definition,
∞ 1
2 4 xy 6 y 2 y =1 4 x + 3
g ( x) =∫ f ( x, y )dy =∫ (2 x + 3 y )dy =( + ) |y=0 = ,
−∞ 0 5 5 10 5
for 0 ≤ x ≤ 1, and g(x) = 0 elsewhere. Similarly,
∞ 1
2 2(1 + 3 y )
h( y ) = ∫ f ( x, y )dx = ∫ (2 x + 3 y )dx = ,
−∞ 0 5 5
for 0 ≤ y ≤ 1, and h(y) = 0 elsewhere.

Example 12.6: Suppose that the shelf life, in years, of a certain perishable food product packaged
in cardboard containers is a random variable whose probability density function is
 e− x , x > 0,
given by f(x) = 
 0, elsewhere.
Let X 1 , X 2 , and X 3 represent the shelf lives for three of these containers selected
independently and find P(X 1 < 2, 1 < X 2 < 3, X 3 > 2).

PHM111s - Probability and Statistics

Solution: Since the containers were selected independently, we can assume that the random
variables X 1 , X 2 , and X 3 are statistically independent, having the joint probability density
f(x 1 , x 2 , x 3 ) = f(x 1 )f(x 2 )f(x 3 ) = e − x e − x e − x = e − x − x − x ,
1 2 3 1 2 3

for x 1 > 0, x 2 > 0, x 3 > 0, and f(x 1 , x 2 , x 3 ) = 0 elsewhere. Hence

∞ 3 2

P(X 1 < 2, 1 < X 2 < 3,X 3 > 2) = ∫ ∫ ∫ e − x − x − x dx1dx2 dx3

1 2 3

2 1 0

= (1 − e −2 )(e −1 − e −3 )e −2 =
0.0372.
Definition: Let X and Y be random variables with joint probability distribution f(x, y). The mean,
or expected value, of the random variable g(X, Y) is
∞ ∞

μ g(X,Y) = E[g(X, Y)] = ∫ ∫ g ( x, y ) f ( x, y )dxdy if X and Y are continuous.

−∞ −∞

Note that if g(X, Y) = X , we have

∞ ∞ ∞

E(X) = ∫ ∫ xf ( x, y )dydx = ∫ xg ( x)dx (continuous case),

−∞ −∞ −∞

where g(x) is the marginal distribution of X.

Similarly, we define
∞ ∞ ∞

E(Y) = ∫∫
−∞ −∞
yf ( x, y )dxdy =
−∞
∫ yh( y )dy (continuous case),

where h(y) is the marginal distribution of the random variable Y.

Definition: Let X and Y be random variables with joint probability distribution f(x, y). The
covariance of X and Y is
∞ ∞

σ XY = E[( X − µ X )(Y − µY )] = ∫ ∫ (x − µ X
)( y − µY ) f ( x, y )dxdy , X and Y are continuous.
−∞ −∞

Theorem: The covariance of two random variables X and Y with means µ X and µY , respectively,
is given by
Cov(X, Y) = σ XY = E ( XY ) − µ X µY = E ( XY ) − E ( X ) E (Y ).
Theorem: Let X and Y be two independent random variables. Then
E(XY) = E(X)E(Y).
Corollary: Let X and Y be two independent random variables. Then σ XY = 0.

Definition: Let X and Y be random variables with covariance σ XY and standard deviations σ X
and σ Y , respectively. The correlation coefficient of X and Y is

σ XY
ρ XY = . −1 ≤ ρ XY ≤ 1
σ Xσ Y

PHM111s - Probability and Statistics

=σX E( X 2 ) − E 2 ( X ) = σ XY E ( XY ) − E ( X ) E (Y ) = σy E (Y 2 ) − E 2 (Y )
Example 12.10: The fraction X of male runners and the fraction Y of female runners who
compete in marathon races are described by the joint density function
 8 xy, 0 ≤ y ≤ x ≤ 1,
f(x, y) = 
 0, elsewhere.
Find the correlation coefficient between X and Y.
Solution: We first compute the marginal density functions. They are
 4 x3 , 0 ≤ x ≤ 1,
g(x) = 
 0, elsewhere.
 4 y (1 − y 2 ), 0 ≤ y ≤ 1,
and h(y) = 
 0, elsewhere.
From these marginal density functions, we compute
1 1
4 8
µ X = E ( X ) = ∫ 4 x 4 dx = and µY = E (Y ) = ∫ 4 y 2 (1 − y 2 )dy = .
0 5 0 15
1 1
2 2 1
E( X 2 ) = ∫0 4 x dx = and E (Y 2 ) = ∫ 4 y 3 (1 − y 2 )dy =
1− = ,
5

3 0 3 3
2 2
2  4 2 1  8 11
σ =−   = and σ Y2 =−   = .
2

3  5  75 3  15  225
X

From the joint density function given above, we have

1 1
4
= ∫=
∫ 8 x y dxdy
2 2
E ( XY ) .
0 y 9
4 4 8 4
Then σ XY =
E ( XY ) − µ X µY = − ( )( ) =− .
9 5 15 225
σ XY −4 / 225 −4
Hence, ρ XY =
= = .
σ Xσ Y (2 / 75)(11 / 225) 66
Theorem: If X and Y are random variables with joint probability distribution f(x, y) and a, b, and
c are constants, then
σ aX2 + bY + c = a 2σ X2 + b 2σ Y2 + 2abσ XY .
Corollary 1: Setting b = 0, we see that
σ= 2
aX + c
a= σ X a 2σ 2 .
2 2

Corollary 2: Setting a = 1 and b = 0, we see that

σ X2=+c
σ= 2
X
σ 2.
Corollary 3: Setting b = 0 and c = 0, we see that
= σ aX2 a= σ X a 2σ 2 .
2 2

Corollary 4: If X and Y are independent random variables, then

σ= 2
aX + bY
a 2σ X2 + b 2σ Y2 .
Corollary 5: If X and Y are independent random variables, then
PHM111s - Probability and Statistics
 σ=
2
aX − bY
a 2σ X2 + b 2σ Y2 .
Example: The continuous random variables X & Y have joint density function given by:
C (6 − x − y ), 0 < x < 2 & 2 < y < 4,
f(x, y) = 
 0, elsewhere.
(a) Calculate C.
(b) P( X + Y > 3)
(c) Find the marginal distribution of X alone and of Y alone g(x) and h(y).
(d) Check independence of X & Y.
(e) Find Cov(2X, 5Y).
(f) Find the correlation coefficient between X and Y.
∞ ∞ 4 2

Solution: (a) ∫∫
−∞ −∞
=
f ( x, y )dxdy ∫ ∫ C (6 − x − y )dxdy
2 0
= 1
4
x2
⇒ C ∫ (6 x − − xy ) |xx == 02 dy =
1
2 2
4

⇒ C ∫ (10 − 2 y )dy =
1
2

⇒ C[(10 y − y 2 ) | yy == 42 ] =
1
1
⇒ C (20 − 12) =1 ⇒ C =
8
(b) To calculate the probability, we use
P[(X, Y) ∈ A] = P( X + Y > 3)
3 2 4 2
1 1
∫2 3∫− y 8 (6 − x − y )dxdy + ∫3 ∫0 8 (6 − x − y )dxdy
3 2 4 2
1 x 1 x
⇒ ∫ (6 x − − xy ) | dy + ∫ (6 x − − xy ) | dy
= x 2= x 2
x=3− y x=0
2 8 3 2 8 2
−1 3 2 14 19
⇒ ∫
16 2
( y − 8 y + 7)dy + ∫
83
(10 − 2 y )dy =
24
Method 2:
3 3− y
1
P( X + Y > 3) = 1 − ∫ ∫ (6 − x − y )dxdy
2 0 8
3
1 x2 19
= 1 − ∫ (6 x − − xy ) |30− y dy =
2 8 2 24

PHM111s - Probability and Statistics

 (c) By definition,
∞ 4
1 1
g ( x)= ∫
−∞
f ( x, y )dy= ∫ 8 (6 − x − y )dy=
2 4
(3 − x),

for 0 ≤ x ≤ 2, and g(x) = 0 elsewhere. Similarly,

∞ 2
1 1
h( y )= ∫ f ( x, y )dx= ∫ (6 − x − y )dx= (5 − y ),
−∞ 0 8 4
for 2 ≤ y ≤ 4, and h(y) = 0 elsewhere.
1 1 1
(d) f(x, y) = (6 − x − y ) ≠ (3 − x) . (5 − y ) = g(x)h(y)
8 4 4
⇒ X & Y are not independent.
(e) Cov(2X, 5Y) = σ 2 X ,5= Y
E (2 X *5Y ) − µ2 X µ5=Y
10[ E ( XY ) − µ X µY ]
Solution: from the marginal density functions:
1
 (3 − x), 0 ≤ x ≤ 2,
g(x) =  4
 0, elsewhere.
and
 1
 (5 − y ), 2 ≤ y ≤ 4,
h(y) =  4
 0, elsewhere.
we compute σ 2 X ,5=
Y
E (2 X *5Y ) − µ2 X µ5= Y
10[ E ( XY ) − µ X µY ]
2 4
x 5 y 17
µ X = E ( X ) = ∫ (3 − x)dx = and µY = E (Y ) = ∫ (5 − y )dy = .
0 4 6 2 4 6
2
x2 4
y2 25
E ( X ) = ∫ (3 − x)dx = 1 and E (Y ) = ∫ (5 − y )dy =
2 2
,
0 4 2 4 3
2 2
 5  11 25  17  11
σ = 1 −   =and σ Y2 =−   =.
2

 6  36 3  6  36
X

From the joint density function given above, we have

4 2
xy 56
E (=
XY ) ∫∫
2 0 8
(6 − x − y )dxdy
=
24
Then
56 5 17 1
σ XY =
E ( XY ) − µ X µY = − ( )( ) =
− .
24 6 6 36
5
⇒ σ 2 X ,5Y = 10σ XY = − .
18
Hence,

PHM111s - Probability and Statistics

 σ −1 / 36 −1
ρ XY
= =
XY
= . Try ρ 2 X ,5Y !!!!!!!
σ Xσ Y (11 / 36)(11 / 36) 11

PHM111s - Probability and Statistics