CHAPTER ONE

BIVARIATE DISTRIBUTIONS
1.1: Introduction
Let and be two random variables defined on the same sample space . Then
the ordered pair ( ) is called a two-dimensional random variable or a two-
component random vector. The values of the random vector ( ) are denoted
by ( ) , where is the value assumed by and is the value assumed by
Since each of the random variables and takes values on the real line , it
follows that the random vector ( ) takes values in the two-dimensional real
space . That is ( ) is a point in .
As is in the case of one-dimensional random variables, we also classify two-
dimensional random variables as discrete or continuous, depending on the nature
of the component one-dimensional random variables.
1.2: Discrete Joint Distributions
If both and have discrete distributions, then ( ) is said to have a discrete
bivariate distribution. Alternatively, we may say that and have a discrete joint
distribution. The joint probability function of and is defined to be the function
such that for any point ( ) in ,
( ) ( ).
The probability distribution function ( ) satisfies:
i. ( ) for all ( )

ii. ∑ ∑ ( ) .

Example 1.1: Suppose and are discrete random variables, with joint
distribution

( )
( ) {

i. Verify that ( ) is a joint probability distribution.

1
 ii. Hence compute ( )

Solution
i. Clearly ( ) for all .

∑ ∑ ( ) ∑∑ ( )

∑ ( ) ( ) .
Hence ( ) is a joint probability distribution.
ii. ( ) ( ) ( )

,∑ ∑ ( )-

, - .

Example 1.2:
Each of two boxes contain four identical balls marked with numbers 1 to 4. Two
balls are drawn at random, one from each box, simultaneously. Let denote the
maximum of the numbers on the balls drawn and the sum of the numbers on
the balls drawn. Determine the joint probability distribution of and .
Solution
The sample space S of the random experiment described here contains 16 equally
likely points, each occurring with probability . Thus,

( )( )( )( )( )( )( )( )( )( )( )( )
{ }
( )( )( )( )

The values that ( ) can take on are

( )( )( )( )( )( )( )( )( )( ).
Therefore, ( ) is a discrete random vector with a joint probability distribution
function given by

2
( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
( )

Example 1.3: Consider the bivariate function defined by

( ) 2

Find the value of the constant if ( ) is a joint probability distribution

function (p.d.f) of two random variables and .
Solution
∑ ∑ ( ) since ( ) is a joint p.d.f

Thus, ∑ ∑ .

1.2: Continuous Joint Distributions

A two-dimensional random vector ( ) is said to be continuous if each of its
components and is a continuous random variable.
If both and have continuous density functions, then ( ) is said to have a
two-dimensional or bivariate density function. We may also say that and have
a continuous joint density. Let us now assume that the continuous random vector
( ) assumes values in a region in
*( ) + .
Suppose further that this unit probability is continuously distributed over region
. Our aim to obtain an expression for
*( ) +,
where is a subset of the region , defined by
*( ) +,
Where a, b, c, d are real constants.
*( ) + ( )
3
 ∫ ( )

∫ ∫ ( ) .

A two-dimensional or bivariate density function ( ), for the random vector

( ), is a non-negative real valued function, defined on

i. *( ) + ∫ ∫ ( )
ii. ( ) ∫ ∫ ( )
iii. For *( ) + ∬ ( )

The function ( ) defined above is known as the joint probability density

function of and
Example 1.4: Consider the bivariate function defined by
( )
( ) 2

Determine the value of the constant k so that ( ) is the joint density function
(p.d.f) of the two random variables and Hence evaluate the following
probabilities:
i. ( ),
ii. ( ).
Solution

∫ ∫ ( ) since ( ) is a joint p.d.f.

i.e ∫ ∫ ( )

∫ 0 1

∫ ( )

, -

Hence the function ( ) is given by

4
 ( )
( ) {

i. ( ) ∫ ∫ ( )

ii. *( ) )+ ∬ ( ) , where B is the region

defined by

*( ) +

y
y=4

(0,3)

(1,2)
y=2
x + y =3
O
x

5
 ( ) ∫ ∫ ( )

Example 1.5
Let X and Y have the joint density function given by

2 ( )3
( ) {

Determine the value of the constant

Solution
Completing the squares in terms involving y we have

( ) { [. / ]}

∫ ( ) 4 5 ∫ { . / }

Making the change of variable . Thus, we have

∫ { . / } ∫ 2 3 √

Therefore,
√
∫ ( ) √ ∫ . / √
√

Which gives
√
.

Exercises
1. Let and have the joint p.d.f given by

( ) 2

6
 Compute the following probabilities:
i. . /

ii. ( )

iii. ( )

iv. ( )

v. . /

vi. ( )

2. Suppose that the joint p.d.f of and is specified as follows:

( ) {
Determine the value of the constant c and hence find the value of
( )

3. Consider a sample of size 2 drawn without replacement from an urn

containing three balls, numbered 1, 2 and 3. Let be the smaller of the two
numbers drawn and the larger. Find the joint discrete density function of
and