0% found this document useful (0 votes)
289 views12 pages

Bivariate Random Variable

This document defines and provides examples of bivariate random variables. It discusses: 1) A bivariate random variable is a pair of random variables where the individual variables may be related. It can be discrete or continuous. 2) The joint probability mass/density function defines the probability of two variables taking on specific values simultaneously. 3) Marginal functions give the probability of individual variables, regardless of the other. 4) Variables are independent if knowing one provides no information about the other. Examples calculate the joint, marginal, and conditional distributions and check for independence.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
289 views12 pages

Bivariate Random Variable

This document defines and provides examples of bivariate random variables. It discusses: 1) A bivariate random variable is a pair of random variables where the individual variables may be related. It can be discrete or continuous. 2) The joint probability mass/density function defines the probability of two variables taking on specific values simultaneously. 3) Marginal functions give the probability of individual variables, regardless of the other. 4) Variables are independent if knowing one provides no information about the other. Examples calculate the joint, marginal, and conditional distributions and check for independence.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

2.

2
Bivariate random variable
In the real life situations more than one variable effects the outcome of a random
experiment. For example, consider an electronic system consisting of two
components. Suppose the system will fail if both the components fail. The
probability distribution of the life of the system depends jointly on the probability
distributions of lives of the components. Knowing the probability distributions of
lives of the components will not provide us the enough information. What we
need is the probability distribution of the simultaneous behavior of lives of the
components. A pair of random variables is known as a bivariate random variable.
The individual random variables in the pair may be related.

Bivariate random variable: let be the sample space associated with a


random experiment. Let be the real line. If ,
, then the pair is known as a
bivariate random variable.

Note:

1. If and are both discrete random variables, then is a bivariate


discrete random variable.
2. If and are both continuous random variables, then is a bivariate
continuous random variable.

Joint probability mass function: Let be a bivariate discrete random


variable, which takes the values for and .
Let

and

 
m n
Then and and   p xi , y j  1. The function is
i1 j 1
known as joint probability mass function (j.p.m.f) of .
Marginal probability mass functions: Let be a bivariate discrete random
variable with joint probability mass function given by . The marginal
probability mass function of and are given by

 
n
 p xi ,y j for and
j 1

 
m
 p xi ,y j for
i1
respectively.

Note: and are independent if and only if

Conditional probability mass functions: Let be a bivariate discrete random


variable with joint probability mass function given by . The conditional
probability mass function of given and the conditional probability mass
function of given are given by

for and

respectively.

Example 1: A fair coin is tossed three times. Let be a random variable that
takes the value if the first toss is a tail and the value if the first toss is a head
and be a random variable that defines the total number of heads in the three
tosses. Then

i. Determine the joint, marginal and conditional mass functions of


and .
ii. Are and independent?
Solution:

i. The sample space and values of and are given in the following table:

Out comes in sample space Value of Value of

Here takes the values and and takes the values and . Then the
j.p.m.f of is computed as below:

The m.p.m.f of is given by

and
The m.p.m.f of is given by

The conditional p.m.f of given is computed below:

The conditional p.m.f. of given is computed as below:

,
,

ii. Here , and


Since , and are not independent .

Example 2 : The j.p.m.f. of is given by

k  2 x  y  for x = 1, 2; y = 1, 2
p  x, y   

0 otherwise

where is a constant

a. Find the value of


b. Find marginal and conditional p.m.fs.
c. Are and independent.

Solution:
2 2
a. Since is a j.p.m.f,   p  x, y   1
x1 y1

2 2 2 2
  p  x, y   k    2 x  y   k  3  4  5  6   18k  1.
x1 y1 x1 y1

Thus

b. The m.p.m.f. of is given by


2 1 2 1 4x  3
p1  x    p  x, y    2 x  y   2 x  1   2 x  2   
y1 18 y1 18 18

Thus, for

The m.p.m.f of is given by


2 1 2 1 2y  6 y 3
p  y    p  x, y    2 x  y   2  y    4  y   
2 x1 18 x1 18 18 9

Thus, for

The c.p.m.f. of given is given by

Thus, for

The c.p.m.f. of given is given by

Thus, for

c. Note that .
Thus, and are not independent.

Joint probability density function: Let be a bivariate continuous random


variable. Let
bd
P  a  x  b,c  y  d     f  x, y dx dy
ac
for some real numbers such that and . Then

i. and
 
ii.   f  x, y dx dy  1
 

and the function is known as the joint probability density function of the
bivariate continuous random variable .

Marginal probability density function: Let be a bivariate continuous


random variable with j.p.d.f. . The marginal probability density functions of
and are given by
 
 f  x, y dy and  f  x, y dx
 

respectively.

Note: and are independent if and only if

Conditional probability density functions: Let be a bivariate continuous


random variable with j.p.d.f. . Let and be the m.p.d.fs of and
respectively.The conditional probability density function of given and the
conditional probability density function of given are given by

and

respectively.

Cumulative distribution function: The cumulative distribution of a bivariate


random variable is defined by

and
   p  t,s  if  X ,Y  is a d.r.v with j.p.m. f . p  x, y 
t  x s y
F  x, y     
   f  t,s dt ds if  X ,Y  is a c.r.v with j.p.d. f . f  x, y 
 

Properties of cumulative distribution function

1.
2.
3. and

4.

Marginal cumulative distribution function: Let be a bivariate random


variable with c.d.f. .The marginal cumulative distribution functions of
and are given by and respectively.

Note:

If is a bivariate continuous random variable with c.d.f. , then its


j.p.d.f. is given by

Example 3: The j.p.d.f. of is given by

  x y 
f  x, y   e for 0  x  ,0  y  
0 otherwise

a. Find marginal p.d.fs of and .


b. Are and independent?

Solution:

a. The m.p.d.f of is given by


   x y  
f1  x    f  x, y dy   e  dy  e x  e y dy  e x .1  e x
0 0 0
e x for x  0
 
0 otherwise

The m.p.d.f. of is given by


  
 x y 
f 2  y    f  x, y dx   e  dx  e y  e xdx  e y .1  e y
0 0 0

e y for y  0
 
0 otherwise

b. Since and are independent.

Example 4: The j.p.d.f. of is given by

  x y1
f  x, y    xe 0  x   ,0  y  

0 otherwise

a. Determine the marginal and conditional p.d.fs


b. Are and independent.

Solution: The m.p.d.f of is given by

   x y1    xy  y 
f1  x    f  x, y dy   xe dy  xe x  e  xy dy  xe x 
e
  e x
0 0 0  x  y 0

 f1  x   e x for 0  x  

The m.p.d.f. of is given by



  x y1 
 x y1
e  dx
    x y 1
dx     1
x.e
f 2  y    f  x, y dx   xe  
y 1 y 1 0
0 0  
0

(using integration by parts)



1   x y1  1
 0 e  
 y  12   0  y  12

1
 f2  y   for 0  y  
2
 y 1
The conditional p.d.f of given is given by

for

The conditional p.d.f of given is given by

for

Note that . Hence, and are not


independent .
Example 5: The j.p.d.f of is given by for
.

a. Find
b. Find the m.p.d.fs of and
c. Are and independent.

Solution:

3 
21 21 2 1 2 1
a. We have   f  x, y dx dy    kx y dx dy  k  x   y dy dx  k  x3 dx
3
00 00 0 0  0 2

2
k  x4  k
  16  2k
2 4  8
 0

21 1
Now ,   f  x, y dx dy  1  2k  1  k  2
00

The j.p.d.f of is given by

for

The m.p.d.f of is given by

1
1 1 31 1 3  y 2  1
 f  x, y dy  2 x  y dy  2 x 2  x3
  4
0 0  0

for

The m.p.d.f. of is given by

2
2 1 2 1 x  4 
 f  x, y dx  2 y  x dx  2 y  4   2 y
3
 
0 0  0
for
b. Note that

Since , and are independent.

You might also like