RANDOM VARIABLES AND THEIR DISTRIBUTION FUNCTIONS

6.1 Types of Random variables: There are two types of random variables namely,
(1) Discrete random variables and (2) Continuous random variables.
1) Discrete random variable: A random variable which takes only a
finite or countable infinite number values is called a discrete random
variable.
Examples: Test scores, number of students, marks obtained by students, number
of
children is different families, number of heads in two-coin tossing, etc.,
2) Continuous random variable: A random variable which takes all
values in one or more intervals is called a continuous random variable.
Examples: Height, weight, age, length, temperature etc.,
Remarks: It X and Y are random variables then
(i) k X (ii) X+Y (iii) X-Y (iv) k1 X+ k2 Y and (v) X Y are also random
variables. Here k, k1 and k2 are constants.

6.2 Probability Distribution of a Discrete Random variable or Probability Mass

Function:
Suppose X be a discrete random variable which assumes a finite or
countably
infinite number of values say x1, x2 …… ,With each of these values, we
associate a weight (or mass) pi = P(X = xi) , i = 1, 2 …… ,which is known
as the
probability of xi, such that
1) pi  0  i = 1, 2, ….. and

2) p
i 1
i  1 i.e., sum of weights (probabilities) is unity.

The function pi = P(X = xi) is called the probability mass function (p.m.f)
of the
random variable X or the probability function of the random variable X.
Further, the following table that consists of all possible ordered pairs {x,
p(x)}, is called the probability distribution of discrete random variable X;

1
 X: x1, x2 ………………………. xn Total
P(x) p1, p2 ………………………. pn 1

Example: Consider the random experiment of tossing two coins. The sample
space S is
given by
S = HH , HT , TH , TT 
Define a random variable X as the number of heads obtained. Then X
takes
values 0, 1, 2. We have ,
1 2 1
P  X  0  , P  X  1  , P  X  2  
4 4 4
1 2 1
such that P  X  0   P  X  1  P  X  2      1
4 4 4
The probability distribution of X is given by
x 0 1 2 Total
p(x) 1/4 2/4 1/4 1

6.3 PROBABILITY DENSITY FUNCTION OF A CONTINUOUS RANDOM

VARIABLE :
Let X be a continuous random variable so that it assumes all values of
one or
more than one intervals. Consider the infinitesimal interval.
 dx dx 
 x   X  x   and its probability
 2 2 
 dx dx 
P x   X  x
 2 2 

The width of this interval is dx and P stands for the weight assigned to
dx points. Then the weight assigned to one point is the probability density
function (p.d.f) of

2
 the random
variable X, which is denoted by,

P x
dx
 X  x
dx 
 2 2 
f(x) =
dx

PROPERTIES OF PROBABILITY DENSITY FUNCTION:

The probability density function f(x) has the following properties:

1. f(x)  0, - < x < 

2.  f  x  dx  1


b
3. P  a  X  b   f  x  dx
a

The curve y = f(x) is known as the probability curve and the area under the
curve between x = a and x = b refers to the probability that the random
variable X assumes values between a and b. The area under curve
between - and  is unity.

6.5 PROBLEMS ON PROBABILITY DISTRIBUTION OF DISCRETE

RANDOM VARIABLE:

PROBLEM 6.5.1: Three unbiased coins are tossed. X represents the random
variable of
number of heads. Construct its probability distribution.

SOLUTION:
Sample space: S =  HHH  ,  HHT  ,  HTH  , THH  ,  HTT  , THT  , TTH  , TTT 
Define the random variable X as
X = Number of heads.
X = 0, 1, 2, 3
Thus we have
1
P  X  0 = P TTT  
8
3
P  X  1 = P  HTT  ,  THT  ,  TTH  
8

3
 3
p  X  2 = P  HHT  ,  HTH  ,  THH  
8
1
P  X  3 = P  HHH  
8

The probability distribution is given by

x: 0 1 2 3
1 3 3 1
P(x) 8 8 8 8

PROBLEM 6.5.2: Suppose X be a discrete random variable whose probability

distribution is as given below:

x: -2 -1 0 1 2 3
P(x) 0.1 K 0.2 2k 0.3 k
Find the value of K and hence find P  1  X  2

Solution: We have,  P  x  1
 0.1 + k + 0.2 + 2k + 0.3 + k = 1
4k + 0.6 = 1
4k = 1-0.6 = 0.4
k = 0.1
Thus the probability distribution is given by,

x: -2 -1 0 1 2 3
P(x) 0.1 0.1 0.2 0.2 0.3 0.1

Consider P  1  X  2
= P  X  1   1  X  0    X  0    0  X  1   X  1  1  X  2    X  2 

 P  X  1  P  1  X  0  P  X  0  P 0  X  1  P  X  1  P 1  X  2  P  X  2

= 0.1 + 0 + 0.2 + 0 + 0.1 + 0 + 0.3  X is a discrete random variable

 P  1  X  2 = 0.7

4
 Problem 6.5.3.: Given below is the probability distribution of a random variable X:
x: 0 1 2 3
P[X = x] 0.1 0.3 0.5 0.1
2
If Y = X + 2X, then find the probability distribution of Y. Also find
P 0  Y  5
Solution: Given
Y = X2+2X 0 3 8 5
P[Y = y] 0.1 0.3 0.5 0.1

We have, P  0  Y  5 = P Y  0 + P  0  Y  3 + P Y  3 +
P  3  Y  5

= 0.1 + 0 + 0.3 + 0 = 0.4

 P [0 Y  5] = 0.4
Problem 6.5.4: Let P( x) be the probability mass function of a discrete random
variable X
that assumes x1 , x2 , x3 , x4 , such that, 2P( x1 )  3P( x2 )  P( x3 )  5P( x4 ) .
Find the probability distribution.

2
Solution: Given, P  x2  = P  x1 
3
P  x3  = 2 P  x1 
2
P  x4  =P  x1 
5
By definition of p.m.f we have
P  x1   P  x2   P  x3   P  x4   1
2 2
 P  x1   P  x1   2P  x1   P  x1   1
3 5
 2 2
 P  x1  1   2    1
 3 5
61
 P  x1  1
15
15
 P  x1  
16
2 15 10
Thus, P  x2  =  
3 61 61

5
 15 30
P  x3   2  
61 61
2 15 6
P  x4    
5 61 61
The probability distribution is given by:

x x1 x2 x3 x4
15 10 30 6
p ( x) 61 61 61 61

Problem 6.5.5: A lot of 10 items contain 3 defectives out of which 4 items are
drawn at
random. Let X denote the number of the defectives in the sample. What is
the probability distribution of X?

Solution: Given X: Number of defectives in a sample of 4.

4
C0 6 C3 1
We have, P  X  0 = 10

C3 6
4
C1 6 C2 1
P  X  1 = 10

C3 2
4
C2 6 C1 3
P  X  2 = 10
=
C3 10
4
C3 6 C0 1
P  X  3 = 10

C3 30
The probability distribution of X is given b0y

x: 0 1 2 3
1 1 3 1
P(X = 6 2 10 30
x)

6.6: PROBLEMS ON PROBABILITY DENSITY FUNCTION OF CONTINUOUS

RANDOM VARIABLE:

PROBLEM 6.6.1: Examine whether the following is a probability density function

6
 1
f(x) = , a xb
ba
= 0 other wise

SOLUTION: Consider
 a b 


 f  x  dx =  f  x  dx

+  f  x  dx +  f  x  dx
a b

a b b
1
=  0 dx +  b  a dx   0 dx
 a a

b
1
= 0 dx  0
a
ba
1 1
 xa  b  a  1
b
=
b  a  ba
It follows that f(x) is a probability density function
PROBLEM6.6.2: Given below is probability density function of a random variable
X:
f(x) = a e3 x , 0  x   Find the value of a and hence find P  2  X  3

SOLUTION:

1. Since f(x) is probability density function, we have  f  x  dx  1


0 
  f  x  dx   f  x  dx  1
 0
0 
  0 dx   f  x  dx  1
 0

ae
3 x
 dx  1
0

 e3 x 
 a  1
 3  0
 1
 a 0    1
 3
a
  1 or a = 3
3
 f  x   3e3 x , 0  x  

7
 3 3
2. Consider P(2  X  3) =  3 e 3 x
.dx =  3 e3 x .dx
2 2
3
 e3 x   e9 e6  6 9
= 3   3  3  3   e  e
 3  2  
 P  2  X  3 = e6  e9

PROBLEM 6.6.3: Given below is a probability density function

x if 0  x  1

f ( x) =  2  x if 1  x  2
0
 other wise
Compute the probability of the event E = 0.5  X  1.5

Solution: P(E) = P 0.5  X  1.5

1.5 1.0 1.5
=  f  x  dx =  f  x  dx +  f  x  dx
0.5 0.5 1.0
1.0 1.5
=  x dx
0.5
+   2  x  dx
1.0

 x2   x 
1.0
2 1.5

=    2 x  
 2  0.5  2 1.0

1.0  2
 0.5  2
 1.5 2   1.0 2 
=   2 1.5     2 1.0   
2 2  2   2 

 3  3

 

1 1  3 2 2 1
=   2    2
2 8
 2 2
 2
 
1 1 9 3 3
=  3  =
2 8 8 2 4

PROBLEM 6.6.4: The shelf life (in hours) of a certain perishable packaged food is
a
random variable whose probability density function is given by

8
  20, 000
 if x  0
=   x  100 
3
f(x)
 0
 other wise

Find the probability that one of the packages will have a shelf life of (a)
atleast 200 hours; (b) atmost 100 hours; (c) any where from 80 to 120
hours.

SOLUTION: Let X denotes the life of a package food.


20000
a. P  X  200 =   x  100  3
dx
200

 1 
= 20000   
 2  x  100  
2

200


20000  1 
=    x  100  
 
2
2 200

1 10000 1
= 10000  
 300 
2
90000 9

1
 P  X  200  
9
100
20, 000
b. P  0  X  100 =   x  100  3
dx
0

20000  
100
1
=    x  100  
 
2
2 0

 1 
1
= 10, 000   
  200  100  
2 2

= 10, 000
 1  1 
 10000 40000 

= 1   
1 3
 4  4
3
 P  0  X  100 =
4
20, 000  
120 120
20, 000 1
P 80  X  120  ( x  100) dx 
c. =
80
3
2   ( x  100) 2 
80

 1 1
= 10000   
  220  180  
2 2

9
 = 10000 
 1

1  = 100 
 1

1 
 48400 32100 
  484 321 

 P 80  X  120 = 0.1049

PROBLEM 6.6.5: A random variable X has the following probability density

function:
1
 if  3  x  3
f(x) = 6
0 other wise

Find (a) P  X  1 and (b) P   3 X  2   5

Solution:
a. We have P  X  1  P  X  1  1
 P X  1 = 1  P  X  1
1
dx
= 1  P  1  X  1  1  
1
6
1 2 2
= 1   P x  1  
3 3 3

(a)Write,  3 X  2  5 = 3 X  3 = X  1
 3 

 P  3 X  2  5 = P  X  1 =  f  x  dx =  f  x  dx +  f  x  dx
1 1 3

3  3
1 1 1 1
=  6 dx   0 dx = 6  dx  6  2  3
1 3 1

1
 P  3 X  2  5 =
3
PROBLEM 6.6.6: Suppose X be a continuous random variable with probability
density
function,
3x 2 , 0  x  1
f(x) = 
0, other wise
Find a and b such that (i) P  X  a  = P  X  a  , and (ii) P  X  b  0.05

Solution:

10
  3 x dx =  x 
a

P  X  a =
a
1. We have a
2 3 3

0
0

1
=  3 x 2 dx =  x3   1  a3
1
P  X  a
a
a

given, P  x  a = P  x  a

1 1
 a3 = 1-a3  2a3 = 1  a3   a 1/ 3
2 2
1
=  3 x 2 dx =  x3  = 1-b3
1
2. Consider P  X  b
b
b

Given, P  X  b = 0.05
 1-b3 = 0.05
 b3 = 0.95
 b = (0,.95)1/3

6.7 DISTRIBUTION FUNCTION OF A RANDOM VARIABLE ORCUMULATIVE

PROBABILITY FUNCTION OF A RANDOM VARIABLE

Suppose X be a random variable with range -  < x < . For a real value
x, the distribution function of random variable X is denoted by F(x) and is
defined as
F(x) = P(X  x) , -  < x < .
Thus, F(x) gives the cumulative probability that X takes the values from its
lower limit to x.

6.7.1 Properties of Distribution Function:

The following are some important properties of distribution function of a
random variable:

1. 0  F(x)  1, for all real values x

Proof: 0  .P  X  x   1  0  F  x   1
2. For any reals (a, b) such that a< b,
P  a  X  b   F b   F  a 
Proof: Consider  X  b    X  a    a  X  b 
 P  X  b = P  x  a   P a  x  b 
P  X  b  F b
11 P  X  a  F  a 
  P  a  X  b = P  X  b  P  X  a 
 P  a  X  b  = F  b   F b 
3. F(x) is monotonically non-decreasing.
Proof: For x1 < x2 , consider the event  x1  X  x2 
We have, F  x2   F  x1  = P  x1  X  x2   0
 F  x2   F  x1   0
 F  x2   F  x1  or F  x1   F  x2 
Thus for x1 < x2, we have, F(x1)  F(x2) and hence F(x) is a monotonically
non decreasing function.

4. F(-) = 0
Proof: F     P  X    0
Where F     Lt F ( x)
x 

5. F(+) = 1
Proof: F     P  x    1, where F     Lt F ( x) ,
x 

6.7.2 Distribution Functions of a discrete random variable:

Suppose X be a discrete random variable which takes a countable number
of points x1, x2 …. with probabilities pi = P  X  xi  , i = 1, 2, …., such that

 p   P X  x   1
i i

The discrete distribution function of X, say F(x) is defined as.

F  x   P  X  x    P  X  xi 
xi  x

Remarks:
1. The distribution function of discrete random variable is a step function or a
jump function, the length of its jump at X = xi being P  X  xi   pi
2. Let X be a discrete random variable taking values x1, x2, ……. such that
x1 < x2 < …. then the probability mass function can be determine by the
distribution function as follows :
P  X  xi   F  xi   F  xi 1  or P  X  xi   P  X  xi   P  X  xi 1 

6.7.3 Distribution Function of a continuous random variable:

Suppose X be continuous random variable whose probability density
function (p.d.f) is given by f(x). The distribution function of X is defined as

12
 x

F  x   P  X  x =  f  x  dx , where -  < x < 



Remarks :
b
1. P  a  X  b = P  a  X  b = P  a  X  b = P  a  X  b =  f  x  dx
a

2. If F(x) is derivable at x, then the p.d.f. of X can be obtained from F(x) as

d
F '  x  F  x  f  x
dx

6.7 Problems on discrete distribution function

Problem 6.8.1. : An unbiased coin is tossed thrice. Find the distribution function
of the
random variable that represents number of heads and sketch
its graph

Solution: Sample space is given by

S  HHH , HHT , HTH , THH , HTT , THT , TTH , TTT 
Define X = No. of heads. By the definition of distribution function of X
we have,
F(x) = P[X = 0] + P[X = 1] + …. + P[X= xj] , where xj, is the largest value
assumed by
X that is less than or equal to x.
1
 F(0) = F[X = 0] =
8
1 3 4 1
F(1) = P[X=0] + P[X =1] =   
8 8 8 2
1 3 3 7
F(2) = P[X=0] + P[X =1]+ P [X = 2] =   
8 8 8 8
1 3 3 1 8
F(3) = P[X=0] + P[X =1]+ P = [X = 2] + P = [X = 3] =      1
8 8 8 8 8

Hence, the distribution function is given by

0 if x  0
1
 if 0  x  1
F ( x)   8
 1 if 2  x  3
2 13
1 if x  3

Problem 6.8.2 : Let X be a random variable whose probability mass function is
given by,

, x = 1,2,3,4,5. Find (i) F(x), (ii) P   X  

x 1 5
P[X = x] = X 1
15 
2 2 
x
Solution : Given, P [X=x] = , x  1, 2,3, 4,5
15
By the definition of distribution function, we have,
1
F (1)  P[ X  1] 
15
1 2 3
F (2)  P[ X  1]  P[ X  2]   
15 15 15
1 2 3 6
F (3)  P[ X  1]  P[ X  2]  P[ X  3]    
15 15 15 15
1 2 3 4 10
F (4)  P[ X  1]  P[ X  2]  P[ X  3]  P[ X  4]     
15 15 15 15 15

1 2 3 4 5 15
F (5)  P[ x  1]  P[ X  2]  P[ X  3]  P[ X  4]  P[ X  5]       1
15 15 15 15 15 15
The distribution function of X is,

0 if x  1
1
 if 1  x  2
15
3
 if 2  x  3
15
F ( x)  
 6 if 3  x  4
15
10
 if 4  x  5
15
1 if x  5

P[ X  1]  P[ X  1]  1
1 14
We know that, P[ X  1]  1  P[ X  1]  1  P[ X  1]  1  
15 15
14
i.e., P[ X  1] 
15

14
 Further, 
P ( X  1) 
  1
2
X 
5
2
 
 P 1 X 

5
2

 P[ X  2] 
2
15


 1 5
  2 /15  1
 
P ( X  1)   X 
P
 1
 X 
5
( X  1) 
  2 2
 2 2  P[ X  1] 14 /15 7

Problem 6.8.3 : The probability mass function of a random variable X is zero

except at
the points x=0, 1, 2. At these points, p(0)=3C3, p(1)=4C-10C2 and p(2)=5C-1,
C>0
1. Determine C
2. Compute P[X<2] and P[1<X<2]
3. Find the distribution function and draw its graph
1
4. Find the largest x such that F(x)<
2
1
5. Find the smallest x such that F(x) 
3

Solution:
1. p(0)+p(1)+p(2) = 1
3C3+4C-10C2+5C-1 = 1
3C3-10C2+9C-2 = 0
C=1 is a root
(C-1)(3C2-7C+2) = 0
3C2-7C+2 = 0
(3C-1)(C-2) = 0
1
C= , 1, 2
3
C=1, 2 can not be accepted, since if C=1 or 2, P(0)>1
1
Thus, choose C=
3
1 2 6
p(0)= ; p(1)= ; p(2)=
9 9 9
1 2 3 1
3. P(X<2)=P(X=0)+P(X=1)=   
9 9 9 3

15
 6 2
P(1<X2)=P(X=2)= 
9 3
1
1. F(x)<
2
1
P  X  x 
2
 1
Max.  x : P[ X  x]    Max{0,1}  1
 2
1
F(x)
3
1
P[Xx] 
3
 1
Min.  x : P[ X  x]    min.{1, 2}  1
 3

6.8 PROBLEMS ON DISTRIBUTION FUNCTION OF CONTINUOUS RANDOM

VARIABLES

PROBLEM 6.9.1 : The probability density function of the random variable X is

given
K
 if 0  x  4
by, f ( x)   x
0 else where

find (i) the value of K ; (ii) The expression for F(x)
 1
(iii) P  X   and P( X  1) using F ( x)
 2
Solution: 1
 0 4 



f ( x)dx  

f ( x)dx   f ( x)dx   f ( x)dx  1
0 4
4 4
K
  f ( x)dx   dx
0 0 x

16
 4
 
 x1 2  1
K
1   1  4K  1  K 
  4
 
 2 0
 1
 if 0  x  4
f ( x)   4 x
0 else where

x x
(ii) Let x satisfy   x  0, F ( x)  

f ( x)dx   odx  0


0 x
0<x<4, F(x)= 

f ( x)dx   f ( x)dx
0

x
1
= 0 dx
0 4 x

1 12
F ( x)  x
2
x  4, F ( x)  1
Remark: At x=4, F(4)=1. F(x) continues to be one for all values of x larger than 4.
0 if x  0
 1
1
F(x) =  x 2 if 0  x  4
2
1 if x  4

(iii) 
P X 
1 
P X 
1
P X 
 1
 
2  
2  
2
Remark: If x is a point of continuity of f(x),
x
P[X=x]=  f ( x)dx  0
x

1  1
Since x= is a point of continuity of f(x), P  X    0
2  2

P X 

1

2
P X 


1

2
F  1
2


1

1 1 2 1
 
2 2 2 2

17
 
P X 
1

1
 2
 2 2
1 1
P[ X  1]  1  P[ X  1]  1  F (1)  1  
2 2
1
p[ X  1] 
2

PROBLEM6.9.2: Find the distribution function F(x) of the random variable X whose
probability density function is given by,

 x if 0  x  1

f ( x )   2  x if 1  x  2
0 else where

Evaluate P[0.8<X<1.2], using F(x)

x
Solution: Let x  0,  f ( x)dx  0  F ( x)

x 0 x x
Let 0  x  1, F ( x)  

f ( x)dx   f ( x)dx   f ( x)dx   f ( x)dx
 0 0
x
x2
  xdx 
0
2
0 1 x
Let 1  x  2, F ( x)  

f ( x)dx   f ( x) dx   f ( x)dx
0 1
1 x
 0   xdx   (2  x)dx
0 1

 x2
F ( x)   2x 1
2
0 if x  0
 2
x if 0  x  1
2
F ( x)   2
 x  2x 1 if 1  x  2
 2

1 if x  2

[0.8<X<1.2]=[X<1.2]-[X0.8]
P[0.8<X<1.2]=P[X<1.2] –P[X0.8]

18
 =P[X=1.2]+P[X<1.2]-P[X0.8], since P[X = 1.2] = 0
=P[X1.2] – P[X0.8]
=F(1.2)-F(0.8)
 (1.2) 2   (0.8) 2 
  2(1.2)  1     0.68  0.32
 2   2 
 0.36

PROBLEM 6.9.3 : Given distribution function,

 9
1  if x  3
F(x)=  x 2 find P[X5] and P[X>8]
0 other wise

9 16
Solution: P[X5]=F(5)=1- 
25 25
P[X>8]=1-P[X8]=1-F(8)
 9  9
=1- 1    P[X>8] =
 64  64

6.10 TWO DIMENSIONAL RANDOM VARIABLE:

Suppose X and Y be two random variables defined on the same sample space
(S). With every w  S we can associate Z  w   X  w , Y  w  , Z  w  R 2 . Under
certain conditions Z is called a two dimensional random variable.

6.11 Joint, Marginal and Conditional Discrete Probability Distribution or

Bivariate
Discrete Probability Distributions
Suppose X and Y be two discrete random variables defined on the same sample
space S. Let X takes ‘m’ values say x1, x2, ….. xm; and Y takes ‘n’ values say y1,
y2, ……. yn. We have, a two dimensional discrete random variable (X, Y) which
takes ordered pairs of values (x1, y1), (x1, y2), …., (x1, yn), (x2, y1) , ….. (x2, yn), ….
, (xm, yn) the probability of ordered pair (xi, yi) is given by
pij  P  X  xi and Y  yi  Generally pij is known as the joint probability function of
X and Y . Thus the joint discrete probability distribution of (X, Y) can be
represented in the following table.

Joint probability Function of X and Y

Y

19
 X y1 y2 … yj …. yn Total
…. …. p 1j  p1.
x1 p11 p12 p1j p1n j

…. …. p 2j  p2
x2 p21 p22 p2j p2n j

. . . . . . . .
. . . . . . . .
. . . . . . . .

…. …. p ij  pi
xi pi1 pi2 pij pin j

. . . . . . . .
. . . . . . . .
. . . . . . . .

…. …. p mj  pm
xm pm1 pm2 pmj pmn j

Tota p i1  p1  pi 2  p2

i
…. p im  pm …. p in  pn
l i i i 1

a). Marginal probability distribution of X:

The sum of the probabilities of X taking a particular value xi and y taking all
possible values, is called the Marginal probability function of X, which is given by
P  X  xi   P  X  xi , y  y1   P  X  xi , Y  y2   ...  P  X  xi , Y  yn 
= Pi1  Pi 2  ...  Pin = P
i
ij  Pi

Thus, the pairs of values  xi , P  X  xi  give the Marginal probability Distribution
of X;

X x1 x2 ………………. xm
P(X = xi) p1 p2 ……………….. pm

a) Marginal probability Distribution of Y:

20
 The sum of probabilities of Y taking a particular value yj and X taking all possible
values, is called the Marginal Probability Function of Y, which is given by
P Y  y j  = P  X  x1 , Y  y j   P  X  x2 , Y  y j   ...  P  X  xm , Y  y j 
= P1 j  P2 j  ......  Pmj   Pij  P. j
i

Thus, the pairs of values  y j , P Y  y j  give the Marginal Probability Distribution
of Y

Y y1 y2 …………… Yn
P(Y = yi) p.1 p.2 …………… p.n

b) Conditional Probability Distribution of X given Y

The probability of X taking a value xi, given Y takes a value yj is known as the
conditional probability function of X given Y = yj and is defined as

P  X  xi / Y  y j  =

P X  xi , Y  y j  P ij

P Y  y j  P. j

Similarly, the Conditional probability function of Y given X = xi is defined as

P  X  xi , Y  y j  Pij
P Y  y j / X  xi   
P  X  xi  Pi .
Remarks:
1. i pi.  p1.  p2.  ......  pm.  1
2. j p j  p.1  p.2  ..... p.n  1
p p1 j  p2 j  ......  pmj p. j
3. i p.ijj  p. j

p. j
1

p pi1  pi 2  .....  pin pi.

4. j piji.  pi.

pi.
1

6.11 Joint probability distribution function or bivariate continuous probability

distribution function

21
 Suppose (X, Y) be a two dimensional continuous random variable, then the
joint probability distribution function of (X, Y) is denoted by F(x, y) and is
defined as F ( x, y)  P  X  x, Y  y 
x y

=   f  x, y  dx dy
 

 
where f(x, y)  0 and   f  x, y  dx dy  1
 

6.13 Properties of Joint Probability Distribution Function:

1. The value of F(x, y) lies between 0 and 1
i.e., 0  F(x, y)  1
2. F(-, y) = 0 = F(x, -) and F(, ) = 1
3. For the real number a, b, c and d, we have
P  a  X  b, c  Y  d  = F  b, d   F  a, d   F b, c   F a, c 
4. F(x, y) is a monotonically non-decreasing function.
If x1  x2 and y1  y2 then F  x1 y1   F  x2 y2 

6.14 Marginal Probability Distribution Functions:

For a given joint probability distribution function F(x, y), the marginal
probability
distribution functions of X and Y are given by

(a) FX ( x)  P  X  x, Y    F  x,  


 x 

=    f  x, y  dy  dx

  


FY ( y)  P  X  , Y  y   F  , y  or


 y


=    f  x, y  dx  dy

  

Remarks: Let X and Y be the jointly distributed discrete random variables, then the
Marginal probability distribution functions of X and Y are respectively given by,
a) FX ( x)   P  X  x, Y  y 
y

b) FY ( y)   P  X  x, Y  y 
x

22
 6.15 Conditional probability distribution function:

Let (X, Y) be a two dimensional random variable whose joint probability

distribution function is given by F(x, y). The conditional probability
distribution function of Y given X = x, can be denoted by F(y/x) and is
defined as
F(y/x) = P[Y  y/X= x].
Now, the probability distribution function F(x, y) may be expressed in
terms of the conditional distribution function F(y/x) as
x
F(x, y) =  F  y / x d

2 FX  x 

x
=  F ( y / x) f

2 X ( x) dx

dFX ( x)
where  f X ( x) is the marginal p.d.f. of X.
dx
Similarly, we have
y y

F(x, y) =  F  x / y  d F  y    F  x / y  f  y  dy

1 Y

1 Y

dFY ( y )
where  fY ( y ) is the marginal p.d.f. of Y.
dy
Here, F(x/y) = P  X  x / Y  y 
F1  x / y  , F2  y / x  are respectively the conditional distribution function of X
and Y.

6.16 Joint Probability Density Function:

Suppose (X, Y) be a two dimensional continuous random variable, whose joint

probability distribution function is given by F(x, y). If f(x, y) is any continuous
function of x and y, for -  < x < , -  < y < , such that
 2 F  x, y  y x

f  x, y   or F  x, y     f  x, y  dx dy
x y  

then f(x, y) is called the joint probability density function of (X, Y). It should be
 
noted that f(x, y) > 0 and   f  x, y  dx dy  1
 

23
6.17 Marginal probability density function:

The marginal probability density functions of X and Y can be obtained by

differentiating their corresponding Marginal distribution functions as
follows:
d FX  x  
1. fX(x) =   f  x, y  dy
dx 

d FY  y  
2. fY(y) =
dy
  f  x, y  dx


6.18 Conditional probability Density Functions

Suppose f(x, y) be the joint probability density function of a two

dimensional random variable (X, Y); and f X(x) and fY(y) be the marginal
probability density function of X and Y respectively. Then the conditional
probability density function of X given Y = y, is
f  x, y 
f(x/y) = , provided fY(y)  0
fY  y 
Similarly, the conditional probability density function of Y given X = x is
defined as
f  x, y 
f(y/x) = , provided fX (x) 0
fX  x
Remarks: We have,

(i) f(x/y) = F  x / y
x

(ii) f(y/x) = F  y / x
y

6.19 Stochastic Independence of Random variables :

Two random variable X and Y (either continuous or discrete) are said to be
stochastically Independent or independent random variables if their joint
probability density function (or joint probability mass function) can be expressed
as the product of their marginal probability density functions (or probability
mass functions)
Symbolically, we write it as
f(x, y) = fX (x) . fY(y), if X and Y are continuous random variables,
P  X  x, Y  y   P  X  x  P Y  y  if X and Y are discrete random variables

24
 Remark: Stochastic independence of random variables
 F(x, y) = FX(x) . FY(y)

6.20. Problems on discrete probability distribution functions

Problem 6.20.1: In the experiment of rolling two dice, two random variables X and Y
are
defined as follows:
X: Sum of the scores on the upper most faces of the dice
Y: Product of the scores on the upper most faces of the dice.
Find the probability that X=5 and Y=6

Solution: Consider the event : [X=5, Y=6]

This event occurs if either (2, 3) or (3, 2) occurs
2 1
P [X=5, Y=6] = 
36 18

Problem 6.20.2: Given below is the joint probability distribution of two random variables
X and
Y:
X 1 2 3
Y
1 1/12 1/6 0
2 0 1/9
1/5
3 1/18 1/4
2/15
Derive the conditional distribution of Y given X=2.

Solution: P[X=2]=P[X=2, Y=1]+P[X=2, Y=2]+P[X=3, Y=2]

1 1 1 19
=   
6 9 4 36
P[ X  2, Y  1] 1/ 6 6
P[Y  1/ X  2]   
P[ X  2] 19 / 36 19
P[ X  2, Y  2] 1/ 9 4
P[Y  2 / X  2]   
P[ X  2] 19 / 36 19
P[ X  2, Y  3] 1/ 4 9
P[Y  3/ X  2]   
P[ X  2] 19 / 36 19

25
 These conditional probabilities define the conditional probability distribution of Y
given X=3.

Problem 6.20.3 : Let X1 and X2 be two random variables whose joint probability mass
function
is given by,
x x
P[X1=x1, X2=x2] = 1 2 ; x1=1, 2, 3; x2=1, 2
21
Derive the marginal probability distributions

Solution: P[X1=x1]=  P [ X1  x1 , X 2  x2 ]
x2

x1  1 x1  2 2 x1  3
=  
21 21 21
2x  3 2 1  3 5
P[ X1  x1 ]  1 ; P[ X1  1]  
21 21 21
Marginal probability distribution of X1:
X1 1 2
3
P[X1=x1] 5/21 7/21
9/21

P[ X 2  x2 ]   P[ X 1  x1 , X 2  x2 ]   
 x1  x2 

x1 x1  21 
1  x2 2  x2 3  x2
  
21 21 21
3 x2  6
=
21
9
P[ X 2  1] 
21

Marginal probability distribution of X2:

X2 1 2
P[X2=x2] 7/21 12/21

2
Problem 6.20.4: X and Y are discrete random variables, for which, P[X=0, Y=0]= ;
9
P[X=0,

26
 1 1 5
Y=1]= : P[X=1, Y=0]= ; P[X=1, Y=1]= . Examine whenever X and Y are
9 9 9
independent.

Solution:
P [X=0]= P[X=0, Y=0] + P [X=0, Y=1]
2 1 3
=  
9 9 9
P [X=1]= P [X=1, Y=0] + P [X=1, Y=1]
1 5 6
=  
9 9 9
2 1 3
P [Y=0] = P [X=0, Y=0] + P [X=1, Y=0] = 
9 9 9
1 5 6
P [Y=1] = P[X=0, Y=1] + P[X=1, Y=1] =  
9 9 9
3 3 1 2
P [X=0] P [Y=0] =    = P [X=0, Y=0]
9 9 9 9
X and Y are not independent

Problem 6.20.5 : Two cards are randomly drawn (without replacement) from an
arbitrary deck
of 52 playing cards. X and Y are number of aces obtained in the first draw and
number of aces drawn in both draws. Find. (a). the joint probability distribution of
X and Y; (b) the marginal distribution of X; (c) The conditional distribution of Y
given X=1

Solution: Let X = Number of acess obtained in first draw X=0, 1, 2

Y = Number of aces obtained in both draws
Y=0, 1, 2, 3, 4

(a) Consider [X=0, Y=0]

This event occurs if no ace occurs in either of the two draws
4
C0 48 C2 4
C0 46 C2
P [X=0, Y=0] = 52
 50
C2 C2
4
C0 48 C2 4
C1 46 C1
P (X=0, Y=1) = 52
 50
C2 C2
4
C0 48 C2 4
C2 46 C0
P(X=0, Y=2) = 52
 50
C2 C2

27
P [X = 0, Y = 3] = P [X = 0, Y = 4] = 0
P [X = 1, Y = 0] = 0
4
C1  48C1 C0  47C2
3

P [X = 1, Y = 1] = 52
 50
C2 C2
C1  C1 C1  C1
4 48 3 47

P [X = 1, Y = 2] = 52
 50
C2 C2

C1  C1 C 2  C0
4 48 3 47

P [X = 1, Y = 3] = 52
 52
C2 C2
P [X = 1, Y = 4] = 0

P [X = 2, Y = 0] = 0

P [X = 2, Y = 1] = 0

4
C2 48 C0 2 C0 48 C1
P [X = 2, Y = 2] = 52
 50
C2 C2

4
C2  48C0 2
C1  48C1
P [X = 2, Y = 3] = 52
 50
C2 C2

C2  48C0
4 2
C2  48C0
P [X = 2, Y = 4] = 52
 52
C2 C2

These probabilities specify the joint probability distribution of X and Y

(b) Marginal distribution of X:

4
C0  48C2
P [X=0]= 52
C2
4
C1  48C1
P [X=1]= 52
C2
4
C2  48C0
P [X=2]= 52
C2
(c) The conditional distribution of Y given X = 1:
P [ X  1, Y  0] 0
P [Y = 0 / X = 1] =  0
P [ X  1] P[ X  1]

28
 P[ X  1, Y  1]
P [Y=1/X=1] =
P [ X  1]
4
C1 48 C1 C0  47C2
3 4
C1  48C1 C0  47C2
3

 52
 50 52
= 50
C2 C2 C2 C2
P [ X  1, Y  2] C1  47C1
3

P [Y = 2 / X = 1] = 
P [ X  1] 50
C2
P [ X  1, Y  2] C0  47C0
8

P [Y = 3 / X = 1] = 
P [ X  1] 50
C2
P [X = 4 / X = 1] = 0

Problem 6.20.6 : X and Y are two random variables whose joint probability distribution
is given
by
X 1 2 3
Y
1 0.1 0.1 0.2

2 0.2 0.3 0.1

Find (a) the conditional distribution of X given Y=2; (b) P[X+Y<4]

Solution: P [Y = 2] = 0.2 + 0.3 + 0.1 = 0.6

P  X  1, Y  2 0.2 2
(a) P [X = 1 / Y = 2]   
P Y  2 0.6 6
P  X  2, Y  2 0.3 3
P [X = 2 / Y = 2]   
P Y  2 0.6 6
P  X  3, Y  2 0.1 1
P [X = 3 / Y = 2]   
P Y  2 0.6 6

X 1 2 3
P  X  x Y  2 2 3 1
6 6 6

(b)  X  Y  4   X  1, Y  1   X  1, Y  2   X  2, Y  1
P  X  Y  4  P  X  1, Y  1  P  X  1, Y  2  P  X  2, Y  1
=0.1+0.2+0.1=0.4

29
 P  X  Y  4  0.4

Problem 6.20.7 : Two unbiased dice are rolled. X and Y represent the scores on first
and second
dice respectively, find the probabilities given below:
(a) P  X  Y  6 (b) P  X  Y  10
(c) P X  Y  (d) P  X  Y  6 Y  4
Solution:
 a   X  Y  6   X  1, Y  5   X  2, Y  4   X  3, Y  3   X  4, Y  2   X  5, Y  1
5
P  X  Y  6 
36
b   X  Y  10   X  Y  10   X  Y  11   X  Y  12
  X  4, Y  6   X  5, Y  5   X  6, Y  4   X  5, Y  6   X  6, Y  5   X  6, Y  6
6 1
 P  X  Y  10  
36 6
(c)  X  Y    X  1, Y  1   X  2, Y  2  ........   X  6, Y  6
6 1
P X  Y   
36 6
(d)  X  Y  6 Y  4   X  2, Y  4
1
P X  Y  6 Y  4  P  X  2, Y  4 
36

Problem 6.20.8 : The joint probability mass function of two random variables X and Y
is
specified as follows:
1
P  X  x, Y  y    2 x  y  , x  0,1, 2; y  0,1, 2.
27
Find the conditional distribution Y for given X.

Solution :
Y
0 1 2 P[X=x]
X
1 2 3
0 0 
27 27 27

30
 1 2 3 4 9

27 27 27 27
2 4 5 6 15

27 27 27 27

1
f  x, y  
 2 x  y  , x  0,1, 2; y  0,1, 2
27
Marginal distribution of X:
3 9 15
P  X  0  ; P  X  1  ; P  X  2  ;
27 27 27

Conditional probability P Y  y / X  0 ;
P  X  0, Y  0
P Y  0 / X  0  0
P  X  0
1/ 27 1
P Y  1 X  0  
3/ 27 3
2 / 27 2
P Y  2 X  0  
3/ 27 3

Conditional probabilities : P Y  y X  1
2 / 27 2
P Y  0 X  1  
9 / 27 9
3/ 27 3
P Y  1 X  1  
9 / 27 9
4 / 27 4
P Y  2 X  1  
9 / 27 9

Conditional Probabilities : P Y  y X  2
4 / 27 4
P Y  0 X  2  
15 / 27 15

5 / 27 5
P Y  1 X  2  
15 / 27 15
6 / 27 6
P Y  2 X  2  
15 / 27 15
Conditional distribution of Y for X  x :

31
 X
0 1 2
Y
0 0 1/3 2/3
1 2/9 3/9 4/9
2 4/15 5/15 6/15

Problem 6.20.9: X and Y are two random variables whose joint probability density
function is
given by,

x  y if 0<x<1, 0<y<1
f  x, y   
0 otherwise

Find (a) the joint distribution function (b) marginal density functions and (c)
conditional density functions

Solution : The region of integration is as furnished below :

(a) Let (x, y) belong to region – I

0<x<1
0<y<1

x y x
y 
F  x, y      u  v  du dv      u  v dv  du
0 0 0 0 

y
x
 v2 
x
 y2 
  uv   du   uy   du
0
2 0 0
2

32
 x
 u 2 y y 2u  x 2 y xy 2 xy
      x  y
 2 2  0 2 2 2

xy
F  x, y    x  y  , 0  x  1, 0  y  1
2

Let (x, y) belong to region – II

x  1, 0  y  1
x y 1
y  
x y

F  x, y      u  v  du dv      u  v dv  du      u  v dv  du
0 0 0 0  1 0 

1
 y2 
  uy   du  0
0 
2

1
1
y2   u 2 y uy 2   y y2  y
F  x, y    uy   du       1  y 
0 
2  2 2  0  2 2  2

y
F  x, y  
1  y  , x  1, 0  y  1
2
Let (x, y) belong to region – III

x  1, y  1

x y

1 1 y
 
x 1 y

F  x, y      u  v  du dv      u  v dv    u  v  dv  du      u  v dv   u  v  dv  du
0 0 0 0 1  1 0 1 


1 1

     u  v dv  du  0  0
0 0 

1 1
1
 u2 
1
 1  u2 1  1 1
  uv   du   u   du        1
0
0
2 0 2  2 2 0 2 2

Let (x, y) belong to region – IV

33
 0  x  1, y  1


x 1

F  x, y       u  v dv  du
0 0 

1
 x
v2 
  uv   du
0 
2 0
x
 1
x
 u2 u  x2 x
   u   du      
0 
2  2 2 0 2 2

x
F  x, y   1  x  , if 0  x  1, y  1
2

0 if x  0 or y  0
 xy
  x  y if 0  x  1, 0  y  1
2

y
F  x, y    1  y  if x  1, 0  y  1
2
1 if x  1, y  1
x
 1  x  if 0  x  1, y  1
2


(b) Marginal probability density function of X :

1
1
 y2  1
f1  x     x  y  dy   xy 0     x 
1

0  2 0 2

1
f1  x   x  if 0  x 1
2

Marginal density function of Y :

1
1
 x2  1
f 2  y     x  y  dx      xy 0  y 
1

0  2 0 2

34
 1
f2  y   y  if 0  y 1
2

(c) Conditional density function of X given Y :

f  x, y  x  y
g1  x / y   
f1  y  y ½

2 x  y
 if 0  x  1
g1  x / y    2 y  1
 0
 otherwise

Conditional density function of Y given X :

f  x, y  x  y
g2  y / x   
f2  x  x
1
2

2 x  y
 if 0  y  1
g2  y / x    2 x  1
 0
 otherwise

Problem 6.20.10: The joint probability density function of two random variables X and Y
is
given by,
8 xy, 0  x  y  1
F  x, y   
0 otherwise
 1  1 
Find (a) P  X     Y    ,(b) marginal and conditional distributions
 2  4 

Solution: The region of integration is as shown below:

35
(a) The dotted area is the region of integration.
 1  1 
P  X     Y    is the integral value of the joint probability
 2  4 
density function evaluated over OAB .
The horizontal grid in OAB gives the interval of integration for x as
1
0  x  y, 0  y 
4
 1  1 
1/ 4
y 
P  X     Y    
 2  4     8 xy dx dy
0 0 
1/ 4
y 
  8 y   x dx dy
0 0 
1/ 4
 y2 
 0  2  dy
8 y

1/ 4
1/ 4
 y4 
  4 y dy  4  
3

0  4 0
1

256

 1  1  1
P  X     Y    
 2  4   256

(b) Margined density of X:

36
 The perpendicular grid explains the variation of Y.
x  y 1
The joint density, if integrated w.r.t. y over this range, the marginal density
of X is obtained.
1
1 1
 y 2  8x
f1 ( x)   8 xy dy  8 x  y dy  8 x    (1  x 2 )
x x  2 x 2
f1  x   4 x 1  x 2  , 0  x  1
Marginal density of Y:

The horizontal grid explains the variation of x.

0 x y
y
 x2 
y y

f 2 ( y )   8 xy dx  8 y  x dx  8 y  
0 0  2 0
=4y3
f 2 ( y)  4 y 3 , o  y  1
(c) Conditional density of X given Y:
f ( x, y ) 8 xy 2 x
g1  x / y     2
f 2 ( y) 4 y3 y
2x
g1  x / y   ; 0  x  y, 0  y  1
y2
Conditional density of Y given X:

37
 f ( x, y) 8 xy 2y
g2  y / x    
f1 ( x) 4 x(1  x ) 1  x 2
2

2y
g2  y / x   ; x  y  1, 0  x  1
1  x2

Problem 6.20.11: The joint density of X and Y is given as follows:

2 if x  y  1, x  0 and y  0
f  x, y   
0 otherwise
Find the conditional distribution of Y, given X=x.

Solution:

2 if ( x, y) { x, y  : x  0, y  0, x  y  1

F  x, y   

0 otherwise
Vertical variation:
0  y  1 x
1 x
Density of X: f1 ( x)   2dy  2(1  x)
0

2(1  x), if 0  x  1
f1 ( x)  
0
Conditional density of y given x:
f ( x, y) 2 1
g  y / x   
f1 ( x) 2(1  x) 1  x
1
g  y / x  ;0  y  1  x , 0  x  1
1 x

38
Problem 6.20.12: The joint probability density function of X and Y is given by,
6(1  x  y) if x  0, y  0, x  y  1
f  x, y   
0 otherwise
Examine if X and Y are independently distributed.

Solution: Marginal density function of X:

1 x
f1 ( x)   6(1  x  y)dy  3 1  x 
2

f1 ( x)  3(1  x)2 , 0  x  1
Marginal density function of Y:
f 2 ( y)  3(1  y)2
Product of marginal densities: f1 ( x) f 2 ( y)  9 1  x  1  y 
2 2

 f ( x, y)
X and Y are not independently distributed.

Problem 6.20.13: The following expression stands for a joint probability density
function:
k x  x  y  ,if 0  x  1,  x  y  x

f  x, y    find the value of k.

0 otherwise
1 x
Solution:   kx  x  y  dx dy  1
0 x

1
x 
k    x( x  y )dy dx  1
0  x 
1
k  2 x3dx  1
0

k
1 k  2
2
2 x( x  y) if 0  x  1,  x  y  x
f  x, y   
0 otherwise

39
Problem 6.20.14 : Given below is the distribution function of the random variables X
and Y:
1  e x  e y  e x  y , x  0, y  0
F  x, y   
0 , otherwise
Find P(X+Y>3)

 2 F ( x, y )
Solution:  e x  y
xy
e x  y , x  0, y  0
f  x, y   
0 , otherwise
To find the required probability we use the change of variable technique
Let x  y  v
y u
 x u  v  x  v u
Since x>0, v>u
u v
If x and y are replaced by u and v we obtain,
G(u, v)  ev J
Usually when a transformation of variables takes place, to bring equality between
integrals of the two functions, the transformed function should be multiplied by ,
|J|,called the modulus of Jacobian transformation.
If two variables are involved
x x
u v
J
y y
u v
In the present case, we have,
x  v u
y u
x x y y
 1;  1;  1;  0
u v u v
1 1
J  1
1 0
|J| = 1
G  u, v   e  v

40
  e  v , 0  u  v, 0  v  
G  u, v   
0 , otherwise
Let f2(v) be the marginal density of v.
v u
f 2  v    e v du  e  v  du  ve  v
0 0

v e  v , v  0
f2  v   
0 , otherwise
The desired probability is P  X  Y  3

P  X  Y  3  P  V  3   v e v dv
3


  v e v    e v dv
3
3

 3e  e  4e3  0.1991
3 3

P  X  Y  3  0.1991 .

Problem 6.20.15 : The joint probability density function of X and Y is given by

1
 1  xy  , x  1, y  1
f  x, y    4

0 otherwise
Prove that X2 and Y2 are independently distributed.

Solution: Marginal distribution of X:

1
1 1
f1  x    1  xy  dy 
1
4 2
1
 x 1
f1  x    2

0 otherwise
1
1 1
f2  y   1  xy dy 
1
4 2
1
 , y 1
f2  y    2

0 otherwise
x
1
P  X 2  x   P   x  X  x    2
dx
 x

41
 x
x
   x
 2  x

P  X  x   x
2
1
P Y 2  y   y  2
Consider the event,  X 2  x, Y 2  y    x  X  x ,  y  Y  y 
x  y 1 
P  X 2  x, Y 2  y    4
 1  xy  dy  dx  x y
 x  y 

P  X 2  x, Y 2  y   P  X 2  x  P Y 2  y  (from 1 and 2 )

X2 and Y2 are independently distribution.

42