ProbDistributionsRandomVariables

www.rejinpaul.
com
www.rejinpaul.com
MA6453-PROBABILITY AND
QUEUEING THEORY
UNIT I RANDOM VARIABLES

Dr. V. Valliammal
Department of Applied Mathematics
Sri Venkateswara College of Engineering
Download Useful Materials from Rejinpaul.com

www.rejinpaul.com
www.rejinpaul.com
Random Variables
Random variable
A real variable (X) whose value is determined by the
outcome of a random experiment is called a
random variable.
(e.g) A random experiment consists of two tosses
of a coin. Consider the random experiment which is
the number of heads (0, 1 or2)
Outcome: HH HT TH TT
Value of X: 2 1 1 0

www.rejinpaul.com
www.rejinpaul.com
One Dimensional Random Variable and
Two dimensional Random Variable
A real valued function defined on S and talking
values in R(, ) is called a one-dimensional
random variable.
If the values are ordered pairs of real
numbers, the function is said to be two-
dimensional random variable.
Note: In unit-I random variables and their
probability distributions we restricted
ourselves to one dimensional sample spaces.

www.rejinpaul.com
www.rejinpaul.com
Discrete Random Variable
A random variable x which takes a countable

number of real values is called a discrete
random variable.
(e.g) 1. number of telephone calls per unit time
2. marks obtained in a test
3. number of printing mistakes in each
page of a book

www.rejinpaul.com
www.rejinpaul.com
Probability Mass Function

If X is a discrete random variable taking atmost a
countably infinite number of values x1, x2, .., we
associate a number Pi = P(X = xi) = P(xi), called the
probability mass function of X. The function P(xi)
satisfies the following conditions:
(i) P(xi) 0 i = 1, 2, ,

(ii) P(x ) 1
i
i 1

www.rejinpaul.com
www.rejinpaul.com
Continuous Random Variable
A random variable X is said to be continuous if it

can take all possible values between certain
limits.
(e.g.)1. The length of a time during which a vacuum
tube installed is a continuous random variable .
2. number of scratches on a surface, proportion of
defective parts among 1000 tested,
3. number of transmitted in error.

www.rejinpaul.com
www.rejinpaul.com
Probability Density Function
Consider a small interval (x, x+dx) of length dx.
The function f(x)dx represents the probability that
X falls in the interval (x, x+dx)
i.e., P(x X x+dx) = f(x) dx.
The probability function of a continuous random
variable X is called as probability density function
and it satisfies the following conditions.
(i) f(x) 0 x

(ii) f(x)dx 1


www.rejinpaul.com
www.rejinpaul.com
Distribution Function
The distribution function of a random variable X is
denoted as F(X) and is defined as F(x) = P(X x).
The function is also called as the cumulative
probability function.
x
F(x) P(X x) P(x) when X is discrete
x
x
F(x)dx when X is continuous


www.rejinpaul.com
www.rejinpaul.com
Properties on Cumulative Distribution
1. If a b, F(a) F(b), where a and b are real
quantities.
2. If F is the distribution function of a one-
dimensional random variable X, then 0 F(x) 1.
3. If F is the distribution function of a one
dimensional random variable X, then F() = 0 and
F() = 1.

www.rejinpaul.com
www.rejinpaul.com
Problems
1. If a random variable X takes the values 1, 2, 3, 4
such that 2P(X=1)=3P(X=2)=P(X=3)=5P(X=4). Find
the probability distribution of X.
Solution:
Assume P(X=3) = By the given equation

P(X 1) P(X 2) P(X 4)
2 3 5
For a probability distribution (and mass function)
P(x) = 1
P(1)+P(2)+P(3)+P(4) =1
www.rejinpaul.com
www.rejinpaul.com
61 30
1 1
2 3 5 30 61
15 10 30 6
P( X 1) ; P( X 2) ; P( X 3) ; P( X 4)
61 61 61 61
The probability distribution is given by

X 1 2 3 4
15 10 30 6
p ( x)
61 61 61 61

www.rejinpaul.com
www.rejinpaul.com
2. Let X be a continuous random variable having
the probability density function
2
3, x 1
f ( x) x

0, otherwise
Find the distribution function of x.
Solution:
x
x x
2 1 1
F ( x) f ( x) dx
dx 1
1 1x
3 2
x 1 x
2

www.rejinpaul.com
www.rejinpaul.com
3. A random variable X has the probability density
function f(x) given by
x
f ( x) cx e , x0
0, otherwise
Find the value of c and CDF of X.
Solution: x
F x f ( x) dx
f ( x) dx 1 0
0 x x
cx e dx
x
cx e dx 1 0
x x
0 xe dx
x x 0
c x e e 1 x
0 x x
x e e
c1 1 0
x x
c 1 1 x e e
www.rejinpaul.com
www.rejinpaul.com
4. A continuous random variable X has the
probability density function f(x) given by
x
f ( x) ce , x
Find the value of c and CDF of X.

Solution:
f ( x) dx 1

x
ce dx 1

x
2ce dx 1
0
x
2ce dx 1
0

x
2c e 1
0
2c1 1
1
c
2 Download Useful Materials from Rejinpaul.com
www.rejinpaul.com
www.rejinpaul.com
Case(i ) x 0 Case(ii ) x 0
x
x F x f ( x) dx
F x f ( x) dx
x x
ce dx
x x

ce dx 0 x x x
c e dx c e dx
x x 0
c e dx x
0
x
x
c e c e
0
x x
x c ce c
ce x

c 2 e

1 x
e 1
2 e
x
2 2
1 x
2 e , x0
F ( x) x
1
2 e , x 0
2
www.rejinpaul.com
www.rejinpaul.com
5. A random variable X has the following
probability distribution.
X: 0 1 2 3 4 5 6 7
f(x): 0 k 2k 2k 3k k2 2k2 7k2+k
Find (i) the value of k (ii) p(1.5 < X < 4.5 | X >2) and
(iii) the smallest value of such that p(X) > 1/2.
Solution
(i) P( x) 1
2 2 2
0 k 2k 2k 3k k 2k 7k k 1
2 1
10k 9k 1 0 k 1,
10
1
k 0.1
10
www.rejinpaul.com
A 1.5 X 4.5 2,3,4
www.rejinpaul.com
(ii)
B X 2 3,4,5,6,7
A B 3,4
p A B
p(1.5 X 4.5 | X 2) p A | B
p(3,4)

p B p(3,4,5,6,7)
5
2k 3k 5k 5
10
2 2 2 2 7 7
2k 3k k 2k 7k k 10k 6k
10
(iii) X p(X) F(X)
0 0 0
2 2k = 0.2 0.3
3 2k = 0.2 0.5
4 3k = 0.3 0.8
5 k2=0.01 0.81
6 2k2 = 0.02 0.83
7 7k2+k = 0.17 1.00
From the table for X = 4,5,6,7 p(X) > and the smallest value is 4
Therefore = 4.
www.rejinpaul.com
www.rejinpaul.com
Expectation of a Random Variable
The expectation of a random variable X is denoted
as E(X). It returns a representative value for a
probability distribution.
For a discrete probability distribution
E(X) = x p(x).
For a continuous random variable X which
assumes values in (a, b)
b
E(X) xf(x)dx
a

www.rejinpaul.com
www.rejinpaul.com
Properties on Expectation
1. Expectation of a constant is a constant.
2. E[aX] = aE(X), where a is a constant.
3. E(aX + b) = aE(X) + b, where a and b are
constants.
4. |E(X)| E|X|, for any random variable X.
5. If X Y, E(X) E(Y).

www.rejinpaul.com
www.rejinpaul.com
Variance of a Random Variable

The variance of a Random variable X, which is
represented as V(X) is defined as the expectation
of squares of the derivations from the expected
value.
V(X) = E(X2) (E(X))2
Properties On Variance
1. Variance of a constant is 0
2. Var(aX + b) = a2Var(X), where a is a constant.

www.rejinpaul.com
www.rejinpaul.com
Moments and Other Statistical Constants

Raw Moments
Raw moments about origin
b
r x r f(x)dx
a
Raw moments about any arbitrary value A

b
r (x A)r f(x)dx
a
Central moments
b
r E[X E(X)] (X E(X))r f(x)dx
r

www.rejinpaul.com
www.rejinpaul.com
Relationship between Raw Moments and Central
Moments
1 0 (always)
2 2 12
3 3 321 213
4 4 431 6212 314

www.rejinpaul.com
www.rejinpaul.com
Moment Generating Function (M.G.F)

It is a function which automatically generates the
raw moments. For a random variable X, the
moment generating function is denoted as MX(t)
and is derived as MX(t) = E(etX).
Reason for the name M.G.F
M X (t ) E (e tx )
t 2 X 2 t 3X 2
E 1 tX
2! 3!
t 2X2
E(1) E(tx) E
2!
www.rejinpaul.com
www.rejinpaul.com
t2
1 tE(X) E(X 2 )
2!
t2
1 t1 2
2!
Here 1 = coefficient of t in MX(t)
2 = coefficient of t2
in MX(t)
2!
t2
In general r = coefficient of 2!
in MX(t).

www.rejinpaul.com
www.rejinpaul.com
Problems
1. The p.m.f of a RV X, is given by Find MGF, mean
and variance.
Solution
e
M X t E etX tx
p ( x)

1
etx
x 0 2x
x

et

x 0 2
2 3 4
et et et et
....
2 2 2 2
et et et
2 3
et et
4

1 ..
2 2 2 2 2

et 1 et

2 e t
2 et
1
2
www.rejinpaul.com
www.rejinpaul.com
Differentiating twice with respect to t
t t t t
2 e e e e t
M X t 2e
t 2 t 2
2 e 2 e

2
t t t t t
2 e 2e 2e 2 2 e e t 2t
4e 2e
M X t
t 4 t 3
2 e 2 e

put t = 0 above E( X ) M X 0 2

E X 2 M X 0 6
Variance E X E X
2 2
642

www.rejinpaul.com
www.rejinpaul.com
2. Find MGF of the RV X, whose pdf is given by and
hence find the first four central moments.
Solution

M X t E e tX f x dx
e tx

e tx e x dx
0

e t x dx
0
t x
e

t 0

t
www.rejinpaul.com
www.rejinpaul.com
Expanding in powers of t

2 3
t t t
M X t
1
1 ...
t 1 t

Taking the coefficient we get the raw moments

about origin
E X coefficien t of t 1!
1

2 2 2
E X coefficien t of t 2!

2

E X 3 coefficien t of t 3 3!
6
3

E X 4 coefficien t of t 4 4!
24
4

www.rejinpaul.com
www.rejinpaul.com
and the central moments are
1 0
2
2 2 2C11 1 1
2 1 1 1
2
2 2 2 2

2 3
3 3 3C1 2 1 3C 2 1 1 1
6 2 1 1 1 1 2
3 3
3 2 2 3 3

2 4 4
4 4 4C1 3 1 4C 2 2 1 4C 3 1 1
24 6 1 2 1 1 1 9
4 6 4
4 3 2 2 4 4 4


www.rejinpaul.com
www.rejinpaul.com
1
3. If the MGF of a (discrete) RV X is 5 4e
t find the
distribution of X and p ( X = 5 or 6).
Solution
M X t
1 1

5 4e t 4e t
51
5
1 4e t
2 3
4e t 4e t
1 ...
5 5 5 5

By definition
tX
M X t E e
tx
e p ( x)

t0 t t2
1 e p(0) e p(1) e p(2) ...

www.rejinpaul.com
www.rejinpaul.com
Standard Distributions
Binomial Distribution
Assumptions
1. The random experiment corresponds to only
two possibly outcomes.
2. The number of trials is finite.
3. The trials are independent.
4. The probability of success is a constant from trial
to trial.

www.rejinpaul.com
www.rejinpaul.com
Notations
n number of trials
p probability of success
q probability of failure
X A random variable which represents the
number of successes
Binomial Distribution
A discrete random variable X is said to follow
Binomial distribution if its probability mass
function is
P(x) n Cx p x q n x , x 0,1,2,..., n

www.rejinpaul.com
www.rejinpaul.com
Moment Generating Function
The M.G.F of a Binomial variate is
M X (t) E[e tX ]
n
e tx P(x)
x 0
n
M X (t) e tx n C x p x q n x
x 0
n
n C x (pet ) x q n x
x 0
(pet q) n

www.rejinpaul.com
www.rejinpaul.com
Raw Moments
d
1 M X (t)
dt t 0
d
(pet q) n
dt t 0
n(pet q) n 1 pet
t 0
n(p q) n 1 p np [ p q 1]
Hence mean of Binomial distribution = np

www.rejinpaul.com
www.rejinpaul.com
2
d
2 2
M X (t)
dt t 0
d
[npet (pet q) n 1 ]t 0
dt
np[et (n 1)(pet q) n 2 pet (pet q) n 1 e t ]t 0
np[(n 1)(p q) n 2 p (p q) n 1 ]
np[(n 1)p 1] [ p q 1]
n(n 1)p 2 np
Variance of Binomial Distribution

2 2 1
2
n(n 1)p 2 np n 2 p 2
n 2 p 2 np 2 np n 2 p 2
np(1 p)
npq Download Useful Materials from Rejinpaul.com
www.rejinpaul.com
www.rejinpaul.com
Additive Property of Binomial Distribution
Let X1 follow binomial distribution with parameters
n1 and p1. Let X2 follow Binomial distribution with
parameters n2 and p2. Further let X1 and X2 be
independent.
M X1 (t) (q1 p1e t ) n1
M X 2 (t) (q 2 p 2e t ) n 2
Consider M X1 X 2 (t) M X1 (t)MX 2 (t) [ X1 and X 2 are independent]
(q1 p1e t ) n1 (q 2 p 2e t ) n 2
Which is not of the form (q+pet)n. Hence X1 + X2 is

not a binomial variate.

www.rejinpaul.com
www.rejinpaul.com
But if p1 = p2 = p, then q1 = q2 = q. Equation (1)
becomes
M X1 X 2 (t) (q pet ) n1 (q pet ) n 2
(q pet ) n1 n 2
Which is of the form (q + pet)n.

Hence X1 + X2 follows Binomial distribution when
p1 = p2 = p.
i.e., Binomial distribution has additive property
when p1 = p2 = p.

www.rejinpaul.com
www.rejinpaul.com
Problems Based On Binomial Distribution
1. It has been claimed that in 60 % of all solar heat
installation the utility bill is reduced by atleast one-
third. Accordingly what are the probabilities that
the utility bill will be reduced by atleast one-third
in atleast four of five installation.
Solution Given n = 5, p = 60% = 0.6 and
q =p(1-p = 0.4
x 4) p[ x 4] p[ x 5]
5c 4 (0.6) 4 (0.4) 54 5c5 (0.6) 5 (0.4) 55
0.337

www.rejinpaul.com
www.rejinpaul.com
2. The mean and variance of a binomial variate are
6 and 2 respectively. Find P(X 2).
Solution
Given that E(X) = 6 i.e., np = 6 (1)
and V(X) = 2 npq = 2 (2)
Dividing Equation (2) by Equation (1) gives,

npq 2 1
q
np 6 3
1 2
p 1 q 1
3 3

www.rejinpaul.com
www.rejinpaul.com
2
Use p in Equation (1)
3
2
n 6 n 9
3
x 9 x
2 1
P(x) n C x p x q n x 9 C x
3 3
P(X 1) 1 P(X 1)
1 P(X 0)
0 90
2 1
1 9 C 0
3 3
1

www.rejinpaul.com
www.rejinpaul.com
3. An unbiased die is rolled 10 times. Getting an
outcome greater than 4 in a die is termed as a
success. What is the chance of getting at least 8
successes?
Solution
X : number of successes
p : P (number of success) = P (outcome is greater
C 1
than 4 in a die) C 3
2 1
6 1
1 2
q 1 p 1
3 3
n = number of trials = 10
www.rejinpaul.com
www.rejinpaul.com
x 10 x
1 2
P(x) n C x p x q n x 10 C x , x 0,1,2,...10
3 3
P (atleast 8 successes) = P (X 8)
x 10 x
1 2
10 10
P(x)
x 8 x 8
10 x
C
3 3
8 2 9 1 10 0
1 2 1 2 1 2
10 C8 10 C 9 10 C10
3 3 3 3 3 3
3.37 10 3

www.rejinpaul.com
www.rejinpaul.com
Poisson Distribution
The application of Binomial distribution will be
invalid when n and p 0. Hence in the
presence of the above two conditions we need a
theoretical distribution which overcomes the
drawback of Binomial distribution. The Binomial
distribution tends to poisson distribution when
i. The number of trials is indefinitely large i.e., n
ii. The probability of success is very small i.e., p 0
iii. np is a constant i.e., np = .

www.rejinpaul.com
www.rejinpaul.com
Poisson Distribution pmf
A discrete random variable X is said to follow
Poisson distribution with parameter if its
x
probability mass function is P(x) , x 0,1,2,...,
e
x!

M X (t) E[e ] e tx P(x)
tX
x 0

e x
e tx
x 0 x!

(t ) x
e
x 0 x!
ax a

e t
e e e
x 0 x!
1)
e (e
t

www.rejinpaul.com
www.rejinpaul.com
Raw Moments of Poisson Distribution
d
1 M X (t)
dt t 0
d e t
e e
dt t 0
e t
e e e t
t 0
e e

i.e., mean of Poisson distribution = .

www.rejinpaul.com
www.rejinpaul.com
Second Order Raw Moment
d 2 M X (t)
2
dt 2 t 0
d dM X (t)
.
dt dt t 0
d t t
[e e e e ]t 0
dt

e [e e e e e t ]t 0
e t t t e t
e [e e ]
e e 2 e e
2

www.rejinpaul.com
www.rejinpaul.com
Variance of Poisson distribution

Variance = Second order central moment
2 2 12
2 2
Thus in Poisson distribution mean = Variance = .

www.rejinpaul.com
www.rejinpaul.com
Additive Property of Poisson Distribution
Let X1 and X2 be two independent Poisson variates
with parameter 1 and 2 respectively. Then
1 (e t 1) 2 (e t 1)
M X1 (t) e and M X2 (t) e
Consider the variate X1 + X2
Now
M X1 X 2 (t) M X1 (t)...MX 2 (t) [ X1 and X 2 are independent]
1 (e t 1) 2 (e t 1)
e e
(e t 1)(1 2 )
e
(e t 1)
Which is of the form e . Hence X1 + X2 f ollows
Poisson distribution with parameter 1 + 2. i.e., the
Poisson distribution has additive property.
www.rejinpaul.com
www.rejinpaul.com
Prove that poisson distribution is the limiting case
of Binomial distribution.
(or)
Poisson distribution is a limiting case of Binomial
distribution under the following conditions
(i) n, the no. of trials is indefinitely large, i.e, n
(ii) p, the constant probability of success in each
trial is very small, i.e p 0
(iii) np is inf inite or p and q 1 , is positive real
n n
Solution If X is binomial r.v with parameter n & p,
then
www.rejinpaul.com
www.rejinpaul.com
p ( X x) n c x p x q n x , x 0,1,2,....n
n!
p x (1 p) n x
(n x)! x!
n x
n(n 1)(n 2)....(n ( x 1))(n x)!
x
1
(n x)! x! n n
x
n(n 1)(n 2)....(n ( x 1))
x n
1 1
x! n n n
x
x x 1
n
1 2
x n.n 1 n1 .....n1 1 1
n x! n n n n n
x
1 2 x 1
x n
1 1 .....1 1 1
x! n n n n n

www.rejinpaul.com
www.rejinpaul.com
Taking limit as n on both sides
x
x 1 2 x 1
n
lim p( X x) lim 1 1 .....1 1 1

n n
x! n n n n n
x
x 1 2 x 1
n
lim 1 1 .....1 lim 1 lim 1

x! n n n n n n n n
x e x
(1.1....1) . e .1

, x 0,1,2,.....
x! x!
e x
p ( X x)
x!
, x 0,1,2,..... and it is poisson distbn.
Hence the proof.

www.rejinpaul.com
www.rejinpaul.com
Prove that the sum of two independent poisson
variates is a poisson variate, while the difference is
not a poisson variate.
Solution Let X1 and X2 be independent r.v.s that
follow poisson distbn. with Parameters 1 and 2
respectively.
Let X = X1 + X2

www.rejinpaul.com
www.rejinpaul.com
p ( X n) p ( X 1 X 2 n)
n
pX 1 r . p X 2 n r sin ce X 1 & X 2 are independen t
r 0
nr
e 1 .1 e 2 . 2
n r
.
r 0 r! (n r ) !
.1 1 n !.
n r

1 2 nr
e e . 2
r 0 r ! n ! (n r ) !
e ( 1 2 ) n
n !.

nr
. r

r 0 r ! (n r ) !
1 2
n!
e ( 1 2 ) n
e ( 1 2 )
.n c
nr
r 1
r
2 (1 2 ) n
n! r 0 n!
This is poisson with parameter (1 + 2)

www.rejinpaul.com
www.rejinpaul.com
(ii) Difference is not poisson
Let X = X1 X2
E ( X ) EX 1 X 2
E( X 1 ) E( X 2 )
1 2

E( X 2 ) E X 1 X 2
2
E X X 2 X X
2 2
1 1 1 2
E X E X 2 E X E X
2 2
1 2 1 2
(1 1 ) ( 2 2 ) 2(12 )
2 2
(1 2 ) 2 (1 2 )
(1 2 ) 2 (1 2 )
It is not poisson.
www.rejinpaul.com
www.rejinpaul.com
Prove that poisson distribution is the limiting case
of Binomial distribution.
(or)
Poisson distribution is a limiting case of Binomial
distribution under the following conditions
(i) n, the no. of trials is indefinitely large, i.e, n
(ii) p, the constant probability of success in each
trial is very small, i.e p 0

(iii) np is inf inite or p and q 1 , is positive real
n n
Solution If X is binomial r.v with parameter n & p,

then
www.rejinpaul.com
www.rejinpaul.com
p ( X x) n c x p x q n x , x 0,1,2,....n
n!
p x (1 p) n x
(n x)! x!
n x
n(n 1)(n 2)....(n ( x 1))(n x)!
x
1
(n x)! x! n n
x
n(n 1)(n 2)....(n ( x 1))
x n
1 1
x! n n n
x
x x 1
n
1 2
x n.n 1 n1 .....n1 1 1
n x! n n n n n
x
1 2 x 1
x n
1 1 .....1 1 1
x! n n n n n

www.rejinpaul.com
www.rejinpaul.com
Taking limit as n on both sides
x
x 1 2 x 1
n
lim p( X x) lim 1 1 .....1 1 1

n n
x! n n n n n
x
x 1 2 x 1
n
lim 1 1 .....1 lim 1 lim 1

x! n n n n n n n n
x e x
(1.1....1) . e .1

, x 0,1,2,.....
x! x!
e x
p ( X x)
x!
, x 0,1,2,..... and it is poisson distbn.
Hence the proof.

www.rejinpaul.com
www.rejinpaul.com
Problems Based on Poisson Distribution
1. The no. of monthly breakdowns of a computer is
a r.v. having poisson distbn with mean 1.8. Find the
probability that this computer will function for
a month with only one breakdown.
Solution
e x
p ( X x) , given 1.8
x!
e 1.8 (1.8)1
p( x 1) 0.2975
1!

www.rejinpaul.com
www.rejinpaul.com
2. It is known that the probability of an item
produced by a certain machine will be defective is
0.05. If the produced items are sent to the market
in packets of 20, find the no. of packets containing
atleast, exactly, atmost 2 defectives in a
consignment of 1000 packets using poisson.
Solution
Give n = 20 , p = 0.05 , N = 1000
Mean = np = 1
Let X denote the no. of defectives.

www.rejinpaul.com
www.rejinpaul.com
e . x e 1 .1x e 1 .
p X x x 0,1,2,...
x! x! x!
px 2 1 px 2
1 p( x 0) p( x 1)
e 1 e 1 1
1 1 2 e 0.2642
0! 1!
Therefore, out of 1000 packets , the no. of packets
containing atleast 2 defectives
N . px 2 1000 * 0.2642 264 packets
e 1
(ii) px 2 0.18395
2!
Out of 1000 packets,=N*p[x=2]=184 packets
(iii) px 2 p[ x 0] p[ x 1] p[ x 2]
e 1 e 1 e 1
0.91975
0! 1! 2!
For 1000 packets = 1000*0.91975=920 packets
approximately.
www.rejinpaul.com
www.rejinpaul.com
3. The atoms of radio active element are randomly
disintegrating. If every gram of this element , on average,
emits 3.9 alpha particles per second, what is the
probability during the next second the no. of alpha
particles emitted from 1 gram is
(i) atmost 6 (ii) atleast 2 (iii) atleast 3 and atmost 6?
Solution
Given = 3.9
Let X denote the no. of alpha particles emitted
(i ) p ( x 6) p ( x 0) p ( x 1) p ( x 2) ...... p ( x 6)
e 3.9 (3.9) 0 e 3.9 (3.9)1 e 3.9 (3.9) 2 e 3.9 (3.9) 6
......
0! 1! 2! 6!
0.898

www.rejinpaul.com
www.rejinpaul.com
(ii ) p( x 2) 1 p( x 2)
1 p( x 0) p( x 1)
e 3.9 (3.9) 0 e 3.9 (3.9)1
1
0! 1!
0.901
(iii ) p(3 x 6) p( x 3) p( x 4) p( x 5) p( x 6)
e 3.9 (3.9) 3 e 3.9 (3.9) 4 e 3.9 (3.9) 5 e 3.9 (3.9) 6

3! 4! 5! 6!
0.645

www.rejinpaul.com
www.rejinpaul.com
Geometric Distribution
A discrete random variable X which represents the
number of failures preceding the first, success is
said to follow Geometric distribution if its
probability mass function is
P(x) = qxp, x = 0, 1, ,

M x (t) E[e ] e tx P(x)
tX
x 0

e tx q x p
x 0

p (qe t ) x
x 0
p[1 qe t (qe t ) 2 ]
p[1 qe t ]1
www.rejinpaul.com
www.rejinpaul.com
Raw Moments of Geometric Distribution
d
1 M X (t)
dt t 0
d
p(1 qe t ) 1
dt t 0
p(1)(1 qe t ) 2 (0 qe t )
t 0
pqet (1 qe t ) 2
t 0
pq q
pq(1 q) 2 2

p p

www.rejinpaul.com
www.rejinpaul.com
d 2 M X (t)
2
dt 2 t 0
d
pqet (1 qe t ) 2
dt t 0
2 pq[et (2)(1 qe t ) 3 (0 qe t ) (1 qe t ) 2 e t ]t 0
pq[2q(1 q) 3 (1 q) 2 ]
2q 1
pq 3 2
p p
2q 2 q
2
p p

www.rejinpaul.com
www.rejinpaul.com
Variance of Geometric Distribution
2 2 12
2q 2 q q 2
2 2
p p p
q2 q q q
2 1
p p pp
qqp

p p
q
[ q p 1]
p2

www.rejinpaul.com
www.rejinpaul.com
Expectation of Geometric Distribution
(without using MGF)

E(X) xP(x) xq x p
x 0 x 0
p[0 q 2q 2 ...]
pq[1 2q 3q 2 ...]
pq[1 q] 2
pq q
2
p p

www.rejinpaul.com
www.rejinpaul.com
Variance of Geometric Distribution
V(X) E(X 2 ) [E(X)]2

E[X ] x 2 P(x)
2
x 0

[x(x 1) x]P(x)
x 0

x(x 1)P(x) xP(x)
x 0 x 0

x(x 1)q x p qp
x 0
p[0 0 2q 2 6q 3 8q 4 ...] q
p
2pq2 [1 3q 4q 2 ...] qp
2pq2 [1 q]3 qp
2pq2 q q2 q
3 2 3
p p p p

www.rejinpaul.com
www.rejinpaul.com
V(X) E(X 2 ) [E(X)]2

q2 q q2
2 2 2
p p p
q2 q q q
2 1
p p p p
q q q
1 2 [ p q 1]
p p p

www.rejinpaul.com
www.rejinpaul.com
Establish the memoryless property of geometric distbn.
Solution If X is a discrete r.v. following a geometric distbn.
p( X x) pq x 1 , x 1,2,.....

p( x k ) pq
x k 1
x 1

p q k q k 1 q k 2 ......

p q k 1 q q 2 ..... p q k (1 q) 1
p q k p 1 q k
Now
px m n and x m
px m n x m
p x m
px m n q m n
m q n px n
p x m q
px m n x m px n

www.rejinpaul.com
www.rejinpaul.com
Problem Based on Geometric Distribution
1. Suppose that a trainee soldier shoots a target in an
independent fashion. If the probability that the target
is shot on any one shot is 0.7.
(i) What is the probability that the target would be hit
in 10 th attempt?
(ii) What is the probability that it takes him less than 4
shots?
(iii) What is the probability that it takes him an even
no. of shots?
(iv) What is the average no. of shots needed to hit the
target?
Solution Let X denote the no. of shots needed to hit
the target and X follows geometric distribution with
pmf pX x p q , x 1,2,....
x 1

www.rejinpaul.com
www.rejinpaul.com
(i) px 10 (0.7)(0.3)101 0.0000138
px 4 p( x 1) p( x 2) p( x 3)
(ii)
(0.7)(0.3)11 (0.7)(0.3) 21 (0.7)(0.3) 31
0.973
(iii) px is an even number p( x 2) p( x 4) ........

(0.7)(0.3)2 1 (0.7)(0.3) 4 1 .....

(0.7)(0.3) 1 (0.3)2 (0.3) 4 .......
.......
0.211 (0.3) 2 (0.3) 2
2

0.21 1 (0.3) 2 (0.21) (0.91)
1 1
0.21
0.231
0.91
1 1
(iv) Average no. of shots E(X)
p 0.7
1.4286

www.rejinpaul.com
www.rejinpaul.com
Uniform Distribution
Continuous Uniform Distribution or Rectangular
Distribution
A continuous random variable X defined in the
interval (a,b) is said to follow uniform distribution
if its probability density function is given by
1
, axb
f(x) b a

0, otherwise
Note: and b are the parameters of the distribution.

www.rejinpaul.com
www.rejinpaul.com
Moment Generating Function of a Uniform
Distribution
M X (t) E(e tX )
b
e tx f(x)dx
a
b
1
e tx
dx
a
ba
b
1 e tx

b a t a
1
MGF (e bt e at )
t(b a)

www.rejinpaul.com
www.rejinpaul.com
Raw Moments of Rectangular Distribution
b b
1
r x f(x)dx x
r r
dx
a a
ba
b
1 x r 1

b a r 1 a
1 b r 1 a r 1
r (1)
b a r 1
When r = 1, Equation (1) becomes

1 b2 a 2
1
ba 2
(b a)(b a) b a
Mean
(b a)2 2
www.rejinpaul.com
www.rejinpaul.com
When r = 2, Equation (1) becomes
1 b3 a 3
2
ba 3
1 (b a)(b 2 ab a 2 )

ba 3
b 2 ab a 2

3
Variance of Uniform Distribution

V(X) E(X 2 ) [E(X)]2
2 12
b 2 ab a 2 b a
2

3 2
4b 2 4ab 4a 2 3b 2 6ab 3a 2

12
a 2 b 2 2ab (a b) 2

12 12
www.rejinpaul.com
www.rejinpaul.com
Problems Based on Uniform Distribution
1.Show that for the uniform distribution
1
f ( x) , a x a , the mgf about origin is sinh at .
2a at
Solution: Given f ( x) 21a , a x a
MGF M x (t ) E etx
a
1
e f ( x)dx e tx
tx
dx
a
2a
a
1 e tx
a
1
e dx
tx

2a a 2a t a

1 at
2at

e e at 1
2at
2 sinh at
sinh at
at
sinh at
M x (t )
at

www.rejinpaul.com
www.rejinpaul.com
2. The number of personel computer (pc) sold daily
at a computer world is uniformly distributed
with a minimum of 2000 pc and a maximum of
5000 pc. Find
(1) The probability that daily sales will fall between
2500 and 3000 pc
(2)What is the probability that the computer world
will sell atleast 4000 pcs?
(3) What is the probability that the computer
world will sell exactly 2500 pcs?
Solution Let X~U(a , b) , then the pdf is given by

www.rejinpaul.com
www.rejinpaul.com
1
f ( x) , a xb
ba
1
, 2000 x 5000
5000 2000
1
, 2000 x 5000
3000
3000
p2500 x 3000 f ( x) dx
(1) 2500
x
3000
1 1

3000
dx
2500
3000 3000 2500

1
3000 2500 0.166
3000
5000
(2) px 4000 f ( x) dx
4000
x
5000
1 1
3000
5000
dx
4000
3000 4000

1
5000 4000 0.333
3000
(3) px 2500 0 (i.e) it is particular point, the value is zero.

www.rejinpaul.com
www.rejinpaul.com
3. Starting at 5.00 am every half an hour there is a
flight from San Fransisco airport to Losangles.
Suppose that none of three planes is completely
sold out and that they always have room for
passengers. A person who wants to fly to Losangles
arrive at a random time between 8.45 am and 9.45
am. Find the probability that she waits
(a) Atmost 10 min (b) atleast 15 min
Solution Let X be the uniform r.v. over the interval
(0, 60). Then the pdf is given by
1
f ( x) , a xb
ba
1
, 0 x 60
60
www.rejinpaul.com
www.rejinpaul.com
(a) The passengers will have to wait less than 10
min. if she arrives at the airport
p (5 x 15) p (35 x 45)
15 45
1 1
dx dx
5
60 35
60

1
60
x 15
5

1
60
x 45
35
1

3
(b) The probability that she has to wait atleast 15

min. p (15 x 30) p (45 x 60)
30 60
1 1
dx dx
15
60 45
60
1

60
30

1 60
x 15 60 x 45
1

2
www.rejinpaul.com
www.rejinpaul.com
Exponential Distribution
A continuous random variable X is said to follow
exponential distribution if its pdf is given by
f ( x) e x , x 0
Mean and variance of an Exponential distribution:

Mean E x x f ( x) dx

x e x
dx x e x dx
0 0

xe x e x
2
0
1 1 1
(0 0) 0 2 2


www.rejinpaul.com
www.rejinpaul.com
1
Mean

x
Ex
2 2
f ( x) dx

x e 2 x
dx x 2 e x dx
0 0

x 2 e x 2 xe x 2e x
3
2
0
2 2 2
(0 0 0) 0 0 3 3 2

Variance E x 2 E ( x)
2
2
1
2 2 1 1
2 2 2 2


www.rejinpaul.com
www.rejinpaul.com
Moment Generating Function of an Exponential
Distribution
M X (t) E[e tX ]

e tx f(x)dx
0

e tx e x dx
0

e x ( t) dx
0

e x( t)

( t) 0

[0 1]
t t

www.rejinpaul.com
www.rejinpaul.com
Establish the memory less property of an exponential distribution.
Solution If X is exponentially distributed, then
px s t x s px t for any s, t 0
The pdf of exponential distn is given by

e x , x0
f ( x)
0 , otherwise

p( x k ) f ( x) dx
k

e x
e dx
x

k k

0 e k e x (1)
px s t and x s
px s t x s
px s
px s t e ( s t )
e t px t
px s e s
px s t x s px t for any s, t 0
www.rejinpaul.com
www.rejinpaul.com
Problems based on Exponential Distribution
The time (in hours) required to repair a machine is
exponentially distributed with parameter = 1/2.
(a)What is the probability that the repair time
exceeds 2 hrs ?
(b)What is the conditional probability that a repair
takes atleast 11 hrs given that Its direction
exceeds 8 hrs ?
Solution If X represents the time to repair the
machine, the density function Of X is given by
f ( x) e x , x0
1 x 2
e x0
2

www.rejinpaul.com
www.rejinpaul.com
p ( x 2) f ( x) dx e x
dx
2 2

x
1 e 2
x
1
e 2
dx
2 2 1
2
2 2

0 e 1 0.3679
px 11 x 8 px 3

f ( x) dx e x dx
3 3

x
1 e 2
x
1
e 2
dx
2 2 1
3
2 3

3

3
0 e 2 e 2 0.2231

www.rejinpaul.com
www.rejinpaul.com
Gamma Distribution
A continuous random variable X is said to follow Gamma
distribution with parameter if its probability density
function is e x

x
x0
1
f(x)

0 otherwise
Moment generating function of gamma distribution

M X (t) E[e ] e tx f(x)dx
tX
0

e x x 1
e tx
dx
0

1
0
x(1 t) 1
e x dx
1

(1 t)

(1 t)

www.rejinpaul.com
www.rejinpaul.com
Raw moments of gamma distribution
d d 2 M X (t)
1 M X (t) 2
dt t 0 dt 2 t 0
d d
(1 t) (1 t) 1
dt t 0 dt t 0
(1 t) 1 (1) ( 1)(1 t) 2 (1)

t 0 t 0
(1 t) 1 ( 1) 2
t 0
Variance of gamma distribution

2 2 12
2 2
Hence in gamma distribution mean = variance =

www.rejinpaul.com
www.rejinpaul.com
Variance of gamma distribution
V(X) E(X 2 ) [E(X)]2
2 2
Additive property of Gamma Distribution
Consider two independent gamma variates X1 and X2 with
parameters 1 and 2 respectively.
M X1 (t) (1 t) 1
M X 2 (t) (1 t) 2
M X1 X 2 (t) M X1 (t)MX 2 (t) [ X1 and X 2 are independent]
(1 t) 1 (1 t) 2
(1 t)(1 2 )
Which is of the form (1t). Hence X1 + X2 is also a Gamma variate
with parameter 1+2. Hence Gamma distribution has additive
property.
www.rejinpaul.com
www.rejinpaul.com
Problems Based on Gamma Distribution
1. In a certain city the daily consumption of electric
power in millions of kilowatt hrs can be treated as
1
central gamma distn with
2
, k 3 . If the power
plant has a daily capacity of 12 million kilowatt

hours. What is the probability that the power
supply will be inadequate on any given day.
Solution Let X be the daily consumption of electric
power Then the density function of X is given by

www.rejinpaul.com
www.rejinpaul.com
x
k k 1
e x
f ( x)
k
3
1 31 2 1 2 2
x x
x e x e
8
2

3 2!
x

x 2e 2

16
p[ the power supply is inadequate]=p[x>12]
x

2 2
x e

12
f ( x) dx 12 16 dx
x
1
x 2 e 2 dx
16 3

1
x x x

2 x e
2 2
8 xe 2
16e 2
16 12
0.0625
www.rejinpaul.com
www.rejinpaul.com
2. The daily consumption of milk in a city in excess
of 20,000 liters is approximately distributed as an
1
Gamma distn with parameter 10000 ,k 2 . The city
has a daily stock of 30,000 liters. What is the
probability that the stock is insufficient on a
particular day.
Solution Let X be the daily consumption, so, the r.v.
Y=X-20000. Then f ( y) y e
k k 1 y
k
Y
2
1 21 10000
y
y e
y

10000

ye 10000
2 (10000) 2 1!
y

10000
ye

(10000) 2
www.rejinpaul.com
www.rejinpaul.com
p[ insufficient stock]=p[X>30000]
=p[Y>10000]
y

10000
ye
p[Y 10000]
10000
f ( y ) dy
10000 (10000) 2
dy
y
1

(10000) 2 ye
3
10000
dy
y
2e 1 , By substituti on method , put t
10000
0.7357

www.rejinpaul.com
www.rejinpaul.com
Normal Distribution
The English mathematician De-Moivre, obtained a
continuous distribution as a limiting case of binomial
distribution in the year 1733. This distribution was
named normal distribution. The first person who
made reference to this distribution was Gauss who
used it to study the distribution of errors in
Astronomy.
Probability Density Function
A continuous random variable X is said to follow
normal distribution with parameters (mean) and 2
(variance), it its density function is given by the
probability law:
2
1 x
1
f(x) e 2
, x , and 0
2
www.rejinpaul.com
www.rejinpaul.com
Remarks
A random variable X which follows Normal
distribution with mean and variance 2 is
represented as X ~ N(, 2).
x
If X is a normal variate, is called as a standard
z
normal variate. If X ~ N(, 2), then z ~ N(0, 1).

The pdf of a standard normal variate Z is given by,
1 Z2 /2
(z) e , Z
2

www.rejinpaul.com
www.rejinpaul.com
Moment Generating Function of Normal
Distribution
If X follows normal distribution with mean and
variance 2, its moment generating function is
derived as follows:
M X (t) E[e tX ]

e tx f(x)dx

2
1 x
1
e tx e 2
dx
2
x
Let z dx dz

x z
xz

1
www.rejinpaul.com
www.rejinpaul.com
1 z2
M X (t) e t( z)
e 2
dz
2
1
1 z2
e
t t z
e 2
dz
2

e t 1
t z z 2

2
e 2
dz

e t
1
(z 2 2t z)

2
e 2
dz

e t
1
(z 2 2t z t 2 2 t 2 2 )
M X (t)
2 e

2
dz

e t
1 1 2 2
(z 2 2t z t 2 2 ) t

2 e

2 2
e dz
1
t t 2 2 1
e 2 (z t2

2
e 2
dz

www.rejinpaul.com
www.rejinpaul.com
z t A
dz dA
z A
zA
1
t t 2 2 1
e 2 A2
M X (t)
2 e

2
dA
1 1
A2 A2
2 2
e e
2
dA 1 because
2
is the pdf of a standard normal variate.
1
t t 2 2
M X (t) e 2

www.rejinpaul.com
www.rejinpaul.com
Generating the raw moments using MGF
1
t t 2 2
M X (t) e 2
1 2 2
t
e e t 2
t 2 2 t 2 2
dM X (t) 2
1 e t e 2
(2t) e 2
e t
dt t 0 2
t 0
d 2 M X (t)
2
dt 2 t 0
t 2 2 t 2 2
2 t 2 2
e e (1) e e
2 t 2 t 2
(2t) e te
2 t
2
t t
2 2
t 2 2
2
e e e e
2 t 2
(2t)
2
t 0
2 2

www.rejinpaul.com
www.rejinpaul.com
Problems based on Normal Distribution
1. Find P (z > -1.29)
Solution
0.5
Z 1.29 0
P (z > -1.29) = P(-1.29 < z < 0) + 0.5

= P (0 < z < 1.29) + 0.5
= 0.4015 + 0.5
= 0.9015
www.rejinpaul.com
www.rejinpaul.com
2. Find P (-1 z - 2)
Solution
Z=-2 Z=-1 0
P (-1 z - 2) = P(0 z - 2) P(0 z -1)

= 0.4772 0.3413
= 0.1359

www.rejinpaul.com
www.rejinpaul.com
3. 20% of the observations in the normal
distribution are below 60. 80% of the observations
are below 120. Compute the mean and standard
deviation of the distribution.
Solution Let X be the normal variate
Given that P(X 60) = 0.20
60
P z 0.20 (1)

Also given that P(X 120) = 0.80
120 (2)
P z 0.80

Representing Equations (1) and (2) in the diagram
www.rejinpaul.com
www.rejinpaul.com
0.20 0.3 0.3 0.20

60 0 120

From the diagram P 0 z 120 0.3

120
0.85 [From the tables]

120 0.85 (3)

www.rejinpaul.com
www.rejinpaul.com
Again from the diagram, 60
P z 0 0.3

60
P 0 z 0.3 [By symmetry]

60
0.85 [From tables]

60 0.85
(4)
Adding Equations (3) and (4)
60 1.7 35.29
Putting 35.29 in 120 0.85
90 and
Hence 90 and 35.29

www.rejinpaul.com
www.rejinpaul.com
Summarizing the Unit

Random Variables: 1. Discrete R.V
2. Continous R.V
Standard Distribution
Under Discrete - 1. Binomial Distn
2. Poisson Distn
3. Geometric Distn
Under Continuous- 1. Uniform Distn
2. Exponential Distn
3. Gamma Distn
4. Normal Distn

3) FORMULAE www.rejinpaul.com
www.rejinpaul.com
Sl. No. Discrete random variable Continuous random variable

1.
p(x ) 1
i
i f(x)dx 1

x
2. F(x) P[X x] F(x) P[X x]

f(x)dx

3. Mean E[X] x p(x )
i
i i Mean E[X] xf(x)dx

4. E[X 2 ]
x f(x)dx
x i2 p(x i ) E[X ]
2 2
i

5.
Var(X) E(X 2 ) [E(X)] 2 Var(X) E(X 2 ) [E(X)] 2

6. Moment E[X ] r
x pi
r
i i Moment E[X ] r
x f(x)dx
r
M.G.F. M.G.F
7.

M X (t) E[e tX ] e tx p(x)
M X (t) E[e tX ] e
tx
x
f(x)dx


www.rejinpaul.com
www.rejinpaul.com
4) E(aX + b) = aE(X) + b
5) Var(aX + b) = a2Var(X)
6) Var(aX bY) = a2 Var(X) + b2 Var(Y)
7) Standard Deviation = Var(X)
8) f(x) = F(x)
9) p(X > a) = 1 p(X a)
p(A B)
10)p(A/B) , p(B) 0
p(B)
11) If A and B are independent, then

p(A B) = p(A).p(B).

www.rejinpaul.com
www.rejinpaul.com
(1).P.D.F, M.G.F, Mean and Variance of all the
distributions:
Sl. No. Distribution P.D.F. (P(X = x)) M.G.F Mean Variance
1. Binomial nCxpxqnx (q+pet)n np npq
2. Poisson e x (e t 1)

e
x!

www.rejinpaul.com
www.rejinpaul.com
Sl. Distribution P.D.F. (P(X = x)) M.G.F Mean Variance
No.
3. Geometric qx1p (or) qxp pe t 1 q

1 qe t p p2
4 uniform 1 (b a) 2
, axb e bt e at ab
f(x) b a
(b a)t 12
0, otherwise 2
5 Exponential 1 1
x
f(x) e , x 0, 0
2
0, otherwise
t
f (x)
6 Gamma x
e x 1
1
f(x) ,0 x , 0
() (1 t)
Thank you

ProbDistributionsRandomVariables

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

ProbDistributionsRandomVariables

Hochgeladen von

Copyright:

Verfügbare Formate

www.rejinpaul.

UNIT I RANDOM VARIABLES

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

A random variable x which takes a countable

Download Useful Materials from Rejinpaul.com

Probability Mass Function

Download Useful Materials from Rejinpaul.com

A random variable X is said to be continuous if it

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

The probability distribution is given by

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Find the value of c and CDF of X.

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Variance of a Random Variable

Download Useful Materials from Rejinpaul.com

Moments and Other Statistical Constants

Raw moments about any arbitrary value A

Download Useful Materials from Rejinpaul.com

4 4 431 6212 314

Download Useful Materials from Rejinpaul.com

Moment Generating Function (M.G.F)

Here 1 = coefficient of t in MX(t)

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Taking the coefficient we get the raw moments

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Hence mean of Binomial distribution = np

Download Useful Materials from Rejinpaul.com

Variance of Binomial Distribution

Which is not of the form (q+pet)n. Hence X1 + X2 is

Download Useful Materials from Rejinpaul.com

Which is of the form (q + pet)n.

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Dividing Equation (2) by Equation (1) gives,

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

i.e., mean of Poisson distribution = .

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

Variance of Poisson distribution

Thus in Poisson distribution mean = Variance = .

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com

lim p( X x) lim 1 1 .....1 1 1

lim 1 1 .....1 lim 1 lim 1

Download Useful Materials from Rejinpaul.com

Download Useful Materials from Rejinpaul.com