Department of Statistics & O.R.

Aligarh Muslim University Aligarh

BA/BSc II Semester

Probability and Probability Distributions

(STB 251)

by

Dr. Haseeb Athar

 Unit - II
Two-Dimensional Random Variable
Contents:
• Definition
• Two-Dimensional Discrete Random Variable
• Two-Dimensional Continuous Random Variable
• Joint Probability Mass Function
• Joint Probability Density Function
• Marginal and Conditional Distributions
• Joint and Marginal Distribution Functions

2 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 Two-Dimensional Random Variables
Definition
Let E be an experiment and S a sample space associated with E. Let X = X ( s ) and Y = Y ( s )
be two functions, each assigning a real number to each outcome s  S . Then ( X , Y ) is a two-
dimensional random variable.

X(s)

S
Y(s)

Two-Dimensional Discrete Random Variable

If the set of possible values of two-dimensional random variable ( X , Y ) is finite or countably
infinite. Then, ( X , Y ) is called two-dimensional discrete random variable. The possible
values of ( X , Y ) may be represented as ( xi , y j ); i = 1, 2,..., n; j = 1, 2,..., m .


( X , Y ) = ( xi , y j ), i = 1, 2,..., n; j = 1, 2,...m 
For example, suppose two unbiased dice are rolled together. If X be the number on first die
and Y on second die, then set of possible values of (X, Y) are
( X , Y ) = (1,1)(1, 2)...(6,1)(6,6)
Two-Dimensional Continuous Random Variable
If ( X , Y ) can assume all values in some non-countable set of two-dimensional plane. Then,
( X , Y ) is called two-dimensional continuous random variable. For example, if ( X , Y ) assume
all values in the rectangle ( x, y) | a  x  b, c  y  d  . Then, ( X , Y ) is a two-dimensional
r.v.
Joint Probability Function or Joint Probability Mass Function
Let ( X , Y ) be a two-dimensional discrete r.v. Then, for each pair of possible values
( xi , y j ), i = 1, 2,..., n; j = 1, 2,..., m there associate a real number p ( xi , y j ) representing
P( X = xi , Y = y j ) , is called joint probability function of ( X , Y ) and satisfying the following
conditions:
i) p ( xi , y j )  0 for every ( x, y )
 
ii)   p ( xi , y j ) = 1
j =1 i =1

The set of triplet xi , y j , p( xi , y j ) , i, j = 1, 2,... is called joint probability distribution of

( X ,Y ) .

3 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
The joint probability function of X and Y can be represented in the tabular form as below:

Y y1 y2 yj ym −1 ym Total
X
x1 p11 p12 p1 j p1m−1 p1m p1•

x2 p21 p22 p2 j p2 m −1 p2m p2•

xi pi1 pi 2 pij pim −1 pim pi •

xn−1 pn −11 pn −12 pn −1 j pn −1 m −1 pn −1 m pn −1•

xn pn1 pn 2 pn j pn m −1 pnm pn •

Total p•1 p•2 p• j p• m−1 p• m p•• = 1

In the above table it may be noted that

P( X = xi , Y = y j ) = p( xi , y j ) = pij ; i = 1, 2,..., n; j = 1, 2,..., m
m n n m
pi• =  pij , p• j =  pij and p•• =   pij
j =1 i =1 i =1 j =1

Joint Probability Density Function

Let ( X , Y ) be a two-dimensional continuous r.v. assuming all values in some region R of
Euclidean plane. The joint probability function f is a function that satisfying the following
conditions:
i) f ( x, y )  0 for every ( x, y )
 
ii) − − f ( x, y)dydx = 1

Marginal Probability Functions

Discrete Case:
If joint distribution of two-dimensional discrete random variable ( X , Y ) is given, then
marginal probability distribution of X is given as
P( X = xi ) = p( xi ) = P( X = xi , Y = y1 or X = xi , Y = y2 or...)

=  p ( xi , y j ) = pi•
j =1

Similarly, the marginal distribution of Y is given as

P(Y = y j ) = p( y j ) = P( X = x1, Y = y j or X = x2 , Y = y j or...)

=  p( xi , y j ) = p• j
i =1

4 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
Continuous Case:
Let f ( ) be the joint pdf of the two-dimensional continuous random variable ( X , Y ) . We
define g ( ) and h ( ) , the marginal probability density functions of X and Y respectively are:

g ( x) = − f ( x, y)dy

and h( y) = − f ( x, y)dx

Conditional Distribution Functions

Discrete Case
Let ( X , Y ) be a two-dimensional discrete r.v. with joint probability function p ( x, y ) . Then
conditional distribution of X given Y = y j is given by

P( X = xi  Y = y j )
P( X = xi | Y = y j ) =
P(Y = y j )

P( X = xi , Y = y j ) p( xi , y j )
or P( X = xi | Y = y j ) = =
P(Y = y j ) p( y j )

Similarly, the conditional distribution of Y given X = xi is given by

P( X = xi , Y = y j ) p( xi , y j )
or P (Y = y j | X = xi ) = =
P ( X = xi ) p ( xi )
It may be noted that
  p( xi , yi ) p( y j )
 p( xi | y j ) =  = =1
i =1 i =1 p( y j ) p( y j )
  p ( xi , yi ) p ( xi )
Similarly,  p ( y j | xi ) =  = =1
j =1 j =1 p ( xi ) p ( xi )

Continuous Case
Let ( X , Y ) be a two-dimensional continuous r.v. with joint pdf f ( x, y ) . Then conditional
pdf of X given Y = y is given by
f ( x, y )
g ( x | y) = , h( y )  0.
h( y )
and the conditional pdf of Y given X = x is
f ( x, y )
h( y | x ) = , g ( x)  0.
g ( x)
Note that
 1  h( y )
− g ( x | y)dx = h( y) − f ( x, y)dx = h( y) = 1

5 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
  1  g ( x)
Similarly, − h( y | x)dy = −
f ( x, y )dy = =1
g ( x) g ( x)

Joint Cumulative Distribution Function

Let ( X , Y ) be two-dimensional random variable. The cumulative distribution function (cdf)
or distribution function (df) of the two-dimensional r.v. ( X , Y ) is defined as
F ( x, y ) = P ( X  x, Y  y )
=  p( x, y) : Discrete Case
x y

x y
= − − f ( x, y)dydx : Continuous Case

Marginal Distribution Functions

If F ( x, y ) is the joint distribution function of two dimensional r.v. ( X , Y ) . Then
i) Discrete Case:
The marginal distribution function of X is given by

F ( x) =  F ( x, y) = P( X  x, Y  y)
y y
and the marginal distribution function of Y is given by

F ( y) =  F ( x, y) = P( X  x, Y  y )
x x

ii) Continuous Case:

The marginal distribution function of X is given by
x
 
F ( x) = − − f ( x, y )dy dx 
and the marginal distribution function of Y is given by
y
 
F ( y ) = − − f ( x, y )dx dy 
Independence of Random Variables
Discrete Case:
Let ( X , Y ) be a two-dimensional discrete r.v., then we say X and Y are independent random
variables, if and only if
i) P( X = xi , Y = y j ) = P( X = xi ) P(Y = y j )
or p ( xi , y j ) = p1 ( xi ). p2 ( y j ) for every i, j

ii) p1 ( xi | y j ) = p1 ( xi )
and p2 ( y j | xi ) = p2 ( y j ) for every i, j

p( xi , y j )
Proof: Since p1 ( xi | y j ) =
p2 ( y j )

6 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 p( xi , y j ) = p1( xi | y j ) p2 ( y j )

Now p1 ( xi ) =  p( xi , y j )
j

=  p1 ( xi | y j ) p2 ( y j )
j

= p1 ( xi | y j ) p2 ( y j )
j
 p1 ( xi ) = p1 ( xi | y j ) for every i, j
Similarly,
p2 ( y j ) = p2 ( y j | xi ) for every i, j
Continuous Case
Let ( X , Y ) be a two-dimensional continuous r.v., then we say X and Y are independent
random variables, if and only if
i) f ( x, y ) = g ( x).h( y )  x, y
ii) g ( x | y ) = g ( x)
h ( y | x ) = h ( y )  x, y
Proof: We have
f ( x, y )
g ( x | y) =
h( y )
 f ( x, y ) = g ( x | y ) h ( y )

Now g ( x) = − f ( x, y)dy

= − g ( x | y)h( y)dy

= g ( x | y)− h( y)dy
 g ( x ) = g ( x | y )  x, y
Similarly,
h( x) = h( y | x), x, y
Example 2.27: The joint probability distribution of ( X , Y ) is given in the following table

X 0 1 2
Y
0 3k k 2k
1 k 2k k

Determine the value of k. Then find

i) P ( X  1, Y = 1) , P( X  2) , P(Y = 1) , P( X  2, Y  1)
ii) Marginal distribution of X and Y.
iii) Conditional distribution of Y given X = 1.

7 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
iv) Conditional distribution X given Y = 1.
v) Are X and Y independent?

Solution: We know that

 p( x, y) = 1
 10k = 1 .
1
or k=
10
Therefore, the joint distribution of ( X , Y ) is
X p( y)
0 1 2
Y
0 0.3 0.1 0.2 0.6
1 0.1 0.2 0.1 0.4
p ( x) 0.4 0.3 0.3 1.0

i) P( X  1, Y = 1) = P( X = 0, Y = 1) + P( X = 1, Y = 1) = 0.1 + 0.2 = 0.3

P( X  2) = P( X = 0) + P( X = 1) + P( X = 2) = 0.4 + 0.3 + 0.3 = 1.0
P(Y = 1) = 0.4
P( X  2, Y  1) = P( X = 0, Y  1) + P( X = 1, Y  1)
= P( X = 0, Y = 0) + P( X = 0, Y = 1) + P( X = 1, Y = 0) + P( X = 1, Y = 1)
= 0.3 + 0.1+ 0.1+ 0.2 = 0.7
ii) The marginal distribution of X is defined as
P( X = x) = p( x) =  p( x, y)
y

Thus, from table

X : 0 1 2
P( X = x) : 0.4 0.3 0.3
The marginal distribution of Y is defined as
P(Y = y ) = p( y ) =  p( x, y )
x

Thus, from table

Y : 0 1
P (Y = y ) : 0.6 0.4
iii) The conditional distribution of Y given X is given as
P(Y = y, X = 1)
P(Y | X = 1) =
P( X = 1)

8 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 P( X = 1, Y = 0) 0.1 1
Y = 0 : P(Y = 0 | X = 1) = = =
P( X = 1) 0.3 3
P( X = 1, Y = 1) 0.2 2
Y = 1: P(Y = 1| X = 1) = = =
P( X = 1) 0.3 3
Thus,
Y : 0 1
P (Y = y | X = 1) : 0.6 0.4

iv) Similar as (iii).

v) In the above table we can observe that p( x, y )  p1 ( x) p2 ( y ) for any x, y . Hence X and Y
are not independent.

Practice Problem 2.11: A two dimensional r.v. ( X , Y ) has the bivariate distribution given
by
p( x, y) = k ( x2 + y); x = 0,1, 2,3, y = 0,1
Determine the value of k. Hence find the marginal distribution of X and Y.

Practice Problem 2.12: A two dimensional r.v. ( X , Y ) has joint pmf given by
1
P ( X = x, Y = y ) = (2 x + y ); x = 0,1, 2, y = 0,1, 2
27
Find the conditional distribution of Y given X = x .

Example 2.28: The joint pdf of two dimensional r.v. ( X , Y ) is given as

f ( x, y ) = kxy; 0  x  2, 0  y  2
= 0, elsewhere
Evaluate the value of k and hence obtain
(i) P ( X  1  Y  1)
(ii) P( X  1| Y  1)
(iii) Marginal pdf of X and Y.
(iv) Conditional pdf of X given Y.
(v) Are X and Y independent?
Solution: We know that
 
− − f ( x, y)dydx = 1
This implies
2 2
0 0 k x y dydx = 1
or
2
 2
k 0 x 0 ydy dx = 1 
2
or 2k 0 x dx = 1

9 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
or 4k = 1
1
 k=
4
Hence
1
f ( x, y ) = x y, 0  x  2;0  y  2
4
(i) P( X  1  Y  1) = P( X  1, Y  1)
1 1
= 0 0 f ( x, y )dydx

=
1 1 1

 x  ydy dx
4 0 0

1 1 1
= 0 xdx =
8 16

P( X  1, Y  1)
(ii) P( X  1| Y  1) =
P(Y  1)
2 1
P(Y  1) = 0 0 f ( x, y)dydx

1 2 1
=   xydydx
4 0 0

=
1 2
4
0 
1
x 0 ydy dx 
1 2 1
= 0 xdx =
8 4
Therefore,
P( X  1, Y  1) 1/16 1
P( X  1| Y  1) = = = .
P(Y  1) 1/ 4 4
(iii) Marginal pdf of X is
2
g ( x) = 0 f ( x, y)dy

x 2
=  ydy
4 0
x
= , 0 x2
2
Marginal pdf of Y is
2
h( y) = 0 f ( x, y)dx

y 2
=  xdx
4 0

10 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 y
= , 0 y2
2
(iv) Conditional pdf of X given Y is

f ( x, y )
g ( x | y) =
h( y )
1
xy
4 x
= = , 0  x  2.
y/2 2
(v) Here it can be observed that

f ( x, y ) = g ( x)h( y ) x, y
x
Also g ( x | y ) = g ( x) =
, x  (0, 2) .
2
Hence X and Y are independent.

Example 2.29: If X and Y are two random variables having density function
1
f ( x, y ) = (6 − x − y ); 0  x  2, 2  y  4
8
Find (i) P( X  1  Y  3) (ii) P( X + Y  3) (iii) P ( X  1| Y  3)
Solution:
(i) P( X  1  Y  3) = P( X  1, Y  3)
1 1 3
= 0 2 (6 − x − y )dydx
8
3
=
8

1 1 3− x
(ii) P( X + Y  3) = 0 2 (6 − x − y )dydx
8

 
= 6 0 2 dy dx − 0 x 2 dy dx
1 1 3− x
8
1 3− x
  X +Y = 3

−   ydydx 
1 3− x
0 2 
=
5 X +Y  3
24

1 2 3 5
(iii) P(Y  3) = 0 2 (6 − x − y )dydx =
8 8
Therefore,

11 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 P( X  1  Y  3) 3 / 8 3
P( X  1| Y  3) = = = .
P(Y  3) 5/8 5
Example 2.30: Let the two-dimensional random variable ( X , Y ) has joint pdf

f ( x, y) = 6x2 y, 0  x  1;0  y  1
= 0, elsewhere
3 1
Find (i) P(0  X  ,  Y  2) , (ii) P ( X + Y  1) , (iii) P ( X  Y ) ,
4 3
(iv) P( X  1| Y  2) .
Solution:
3 1 3/4 1 3/4 2
(i) P(0  X  ,  Y  2) = 0 1/3 f ( x, y )dydx + 0 1 f ( x, y )dydx
4 3
3/4 1 3/4 2
= 0 1/3 6 x2 ydydx + 0 1 0 dydx

8 3/4 2 3
= 0
3 x dx =
9 8

1 1− x
(ii) P( X + Y  1) = 0 0 6 x2 ydydx
1− x
1  y2  1
= 0
6 x2   dx =
 2 0 10

1 x
(iii) P( X  Y ) = 0 0 6 x2 ydydx
1
 x
= 60 x 2 0 ydy dx 
x
1 2  y2 
= 6 0
x   dx
 2 0
1 3
= 30 x 4 dx =
5

1 1 1 2
(iv) P( X  1  Y  2) = 0 0 6 x2 ydydx + 0 1 0 dydx = 1
1 2
P(Y  2) = 0 0 f ( x, y)dydx

12 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 1 1 1 2
= 0 0 6 x2 ydydx + 0 1 0 = 1dydx

Thus,
P( X  1  Y  2) 1
P( X  1| Y  2) = =1
P(Y  2) 1
Example 2.31: Let two-dimensional continuous random variable ( X , Y ) has joint pdf
f ( x, y ) = 2, 0  x  y  1
Find marginal density functions of X and Y.
Solution: We know that
1
g ( x) = x f ( x, y)dy

1
= x 2dy

= 2(1 − x), 0  x  1
Similarly
y
h( y) = 0 2dx = 2 y,0  y  1

Example 2.32: Let two-dimensional continuous random variable ( X , Y ) has joint pdf

f ( x, y) = xe− x( y+1) , x  0; y  0
Find marginal and conditional pdf.
Solution: We have

g ( x) = 0 f ( x, y)dy


= x e− x 0 e− x y dy = e− x , x  0.


and h( y) = 0 x e− x( y +1) dx

Let x( y + 1) = z  ( y + 1)dx = dz , then

1     e − x x p −1dx = ( p ) 
h( y ) = 2 0
z e − z dz
( y + 1)  0 
1
= , y0
( y + 1) 2
Now
g ( x | y ) = f ( x, y ).h( y )

= ( y + 1) 2 x e − x ( y +1) , x  0.

and h( y | x ) = f ( x, y ) g ( x )

= x e− x y , y  0.

13 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
Example 2.33: Let the conditional distribution of X given Y is
e− y y x
g ( x | y) = , x = 0,1, 2...
x!
which follows Poisson distribution with parameter y .

The r.v. Y is continuous with marginal pdf h( y ) = e − y , y  0 .

Show that g ( x) = 2−( x+1) , x = 0,1, 2...

Solution: The joint pdf of X and Y is given by
f ( x, y ) = g ( x | y ) h ( y )

yx
= e −2 y ; x = 0,1, 2,..., y  0
x!
Therefore, the marginal density of X is given by

g ( x) = 0 f ( x, y)dy

 yx
= 0 e −2 y dy
x!
1  −2 y x
=  e y dy
x! 0
( x + 1)
= x +1
= 2− ( x +1) , x = 0,1, 2...
x !2

Example 2.34: Let ( X , Y ) are distributed with a constant density inside a square R of side
b. Find f ( x, y ), F ( x, y ) and marginal pdfs.
Solution: The joint pdf of X and Y is given as
1 1 1
f ( x, y) =  = 2 , ( x, y)  R
b b b
= 0, otherwise
The joint cdf is given by

F ( x, y ) = 0, x  0, y  0
x y1 xy
= 0 0 2
dt1dt2 = 2 , 0  x  b; 0  y  b
b b
by y
= 2 = , x  b;0  y  b
b b
xb x
= 2 = , 0  x  b; y  b
b b
= 1, x  b; y  b

14 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
The marginal pdf of X is
b 1 1
g ( x) = 0 2
dy = , x  (0, b)
b b
and the marginal pdf of Y is
b 1 1
h( y) = 0 2
dx = , y  (0, b)
b b
Example 2.35: The joint pdf of a two-dimensional random variable ( X , Y ) is given by
f ( x, y ) = 2, 0  x  1; 0  y  x
= 0, elsewhere
(i) Find marginal density functions of X and Y.
(ii) Find the conditional distribution of Y given X and X given Y.
(iii)Check the independence of X and Y.
Solution: (i) The marginal density function of X and Y respectively are given by
x
g ( x) = 0 2dy = 2 x, 0  x  1
= 0, elsewhere
1
h( y ) = y 2dx = 2(1 − y), 0  y  1
and
= 0, elsewhere
(ii) The conditional distribution of Y given X is
f ( x, y ) 2 1
h( y | x ) = = = , 0  y  x.
g ( x) 2x x
The conditional density function of X given Y is
f ( x, y ) 2 1
g ( x | y) = = = , y  x  1.
h( y ) 2(1 − y ) (1 − y )
(iii) We have
f ( x, y ) = 2, 0  x  1;0  y  x
and g ( x)  h( y ) = 4 x(1 − y )
 f ( x, y )  g ( x )  h ( y )
Hence, X and Y are not independent.

Example 2.36: Joint distribution of X and Y is given by

f ( x, y ) = 4 x y e − ( x + y2 )
2
; x  0, y  0
Test, whether X and Y are independent? Also find the conditional distribution of X given
Y = y.
Solution: The marginal distribution of X is given by

15 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
 
g ( x) = 0 f ( x, y)dy

= 0 4 xye− ( x + y2 )
2
dy

= 4 xe− x 0 ye− y dy
2 2


= 2 xe− x 0 e−t dt  y 2 = t  2 ydy = dt 
2

 

g ( x) = 2 xe − x ; x  0 .
2

Similarly, marginal distribution of Y is

h( y ) = 2 ye − y ; y  0
2

Since, f ( x, y ) = g ( x)  h( y )
Therefore, X and Y are independent.
The conditional distribution of X given Y is
f ( x, y )
= 2 xe− x , x  0.
2
g ( x | y) =
h( y )
Example 2.37: If joint distribution function of X and Y is given by
F ( x, y ) = 1 − e− x − e− y + e− ( x + y ) ; x  0, y  0.
= 0, elsewhere
(i) Find the marginal densities of X and Y.
(ii) Are X and Y independent?
(iii) Find P ( X + Y  1)
Solution: We have
F ( x, y) = 1 − e− x − e− y + e−( x+ y ) ; x  0, y  0.
Therefore,
2
f ( x, y ) = F ( x, y )
x y
  
= 
x  y

1 − e− x − e− y + e−( x+ y ) 



= [e − y − e − ( x + y ) ]
x
f ( x, y ) = e − ( x + y ) , x  0; y  0

= 0, elsewhere
(i) & (ii) The joint density of X and Y can be expressed as
f ( x, y) = e− x  e− y = g ( x)  h( y)

where g ( x) = e− x , x  0

16 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
and h( y) = e− y , y  0.
Thus, X and Y are independent.
(iii)
P( X + Y  1) = 1 − P( X + Y  1)
1 1− x
P( X + Y  1) = 0 0 f ( x, y)dydx
1− x
= 0 e− x {0 e− y dy}dx
1

= 0 e− x {1 − e−(1− x ) }dx
1

= 1 − 2e−1 .
Thus,
P( X + Y  1) = 2e−1 = 0.7358

Example 2.38: If two-dimensional r.v. has joint pdf

1
f ( x, y ) = xy, 0  y  x; 0  x  2
2
Show that X and Y are dependent.
Solution: We have
x
g ( x) = 0 f ( x, y )dy
x x x3
=  ydy = , 0 x2
2 0 4
2
h( y ) = y f ( x, y )dx

y 2 y3
=  xdx = y − , 0 y2
2 y 4
Since f ( x, y )  g ( x)  h( y )
Hence, X and Y are dependent.

Example 2.39: The joint pdf of a two-dimensional r.v. ( X , Y ) is given as

f ( x, y) = e−( x+ y ) , x  0; y  0.
Are X and Y independent?
Find (i) P( X  1) , (ii) P( X  Y | X  2Y ) , (iii) P(1  X + Y  2)
Solution: The joint pdf of X and Y can be expressed as
f ( x, y) = e− x  e− y = g ( x)  h( y) ,

where g ( x) = e− x , x  0

17 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
and h( y) = e− y , y  0.
Thus, X and Y are independent.

(i) P( X  1) = 1 g ( x)dx

= 1 e− x dx = 1/ e

(ii) 
 y
P( X  Y ) =  0
{0 f ( x, y )dx}dy

= 0 e − y {0 e − x dx}dy
y


= 0 e − y (1 − e − y )dy
[ 1 − e − y = t  e − y dy = dt ]
1
= 0 tdt
= 1/ 2

Similarly,
 2y
P( X  2Y ) = 0 {0 f ( x, y )dx}dy

= 0 e− y {0 e − x dx}dy
2y


= 0 e− y (1 − e−2 y )dy
= 2/3
Therefore,
P( X  Y ) 1/ 2
P( X  Y | X  2Y ) = = = 3/ 4
P( X  2Y ) 2 / 3
(iii)
P (1  X + Y  2)
=  f ( x, y )dydx +  f ( x, y )dydx
I II

1 2− x 2 2− x
= 0 1− x f ( x, y)dydx + 1 0 f ( x, y)dydx

2− x 2− x
= 0 e− x {1− x e− y dy}dx + 1 e− x {0 e− y dy}dx
1 2

2 3
= −
e e2
= 0.3298

The above probability can also be evaluated by using the relation

P(1  X + Y  2) = P( X + Y  2) − P( X + Y  1)

18 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India
Example 2.40: Let X and Y be jointly distributed with pdf
1
f ( x, y ) = (1 + xy ), | x | 1,| y | 1
4
= 0, elsewhere
Show that X and Y are not independent but X 2 and Y 2 are independent.
Solution: The marginal pdf of X is given as
1
g ( x) = −1 f ( x, y)dy

1 1 1
= −1
(1 + xy )dy = , − 1  x  1
4 2
Similarly, the marginal pdf of Y is
1 1
h( y ) = −1 f ( x, y )dx = , −1  y  1
2
Since, f ( x, y )  g ( x).h( y ) , therefore X and Y are not independent.
Now
P( X 2  x) = P(| X | x )
x
= − x
g ( x)dx = x

Similarly,
P(Y 2  y) = P(| Y | y ) = y
Also,
P( X 2  x  Y 2  y ) = P(| X | x ,| Y | y )
x y
= − 
x − y
f (u, v)dvdu

1 x y
=   (1 + uv)dvdu
4 − x − y
= x y
This implies
P( X 2  x, Y 2  y) = P( X 2  x) P(Y 2  y)

Hence, X 2 and Y 2 are independent.

19 Lecture notes by Dr. Haseeb Athar, Department of Statistics & O.R., A.M.U., Aligarh, India