Random Variable

Motivation example In an opinion poll, we might decide to ask 50 people whether they agree
or disagree with a certain issue. If we record a 1 for agree and 0 for disagree, the sample

ld
space for this experiment has 250 elements. If we define a variable X=number of 1s recorded
out of 50, we have captured the essence of the problem. Note that the sample space of X
is the set of integers {1, 2, . . . , 50} and is much easier to deal with than the original sample

or
space.

In defining the quantity X, we have defined a mapping (a function) from the original sample
space to a new sample space, usually a set of real numbers. In general, we have the following

W
definition.

Definition of Random Variable A random variable is a function from a sample space S into
the real numbers.
TU
Example 1.4.2 (Random variables)
In some experiments random variables are implicitly used; some examples are these.

Experiment Random variable

Toss two dice X =sum of the numbers
JN

Toss a coin 25 times X =number of heads in 25 tosses

Apply different amounts of
fertilizer to corn plants X =yield/acre

Suppose we have a sample space

ll

S = {s1 , . . . , sn }

with a probability function P and we define a random variable X with range X = {x1 , . . . , xm }.
A

We can define a probability function PX on X in the following way. We will observe X = xi

if and only if the outcome of the random experiment is an sj S such that X(sj ) = xi .
Thus,
PX (X = xi ) = P ({sj S : X(sj ) = xi }). (1)

1
Note PX is an induced probability function on X , defined in terms of the original function
P . Later, we will simply write PX (X = xi ) = P (X = xi ).

Fact The induced probability function defined in (1) defines a legitimate probability function

ld
in that it satisfies the Kolmogorov Axioms.

Proof: CX is finite. Therefore B is the set of all subsets of X . We must verify each of the
three properties of the axioms.

or
(1) If A B then PX (A) = P (xi A {sj S : X(sj ) = xi }) 0 since P is a probability
function.
(2) PX (X ) = P (m
i=1 {sj S : X(sj ) = xi }) = P (S) = 1.

W
(3) If A1 , A2 , . . . B and pairwise disjoint then

PX (
k=1 Ak ) = P (k=1 {xi Ak {sj S : X(sj ) = xi }})

X
X
= P (xi Ak {sj S : X(sj ) = xi } = PX (Ak ),
k=1 k=1
TU
where the second inequality follows from the fact P is a probability function.

A note on notation: Random variables will always be denoted with uppercase letters and
the realized values of the variable will be denoted by the corresponding lowercase letters.
Thus, the random variable X can take the value x.
JN

Example 1.4.3 (Three coin tosses-II) Consider again the experiment of tossing a fair coin
three times independently. Define the random variable X to be the number of heads obtained
in the three tosses. A complete enumeration of the value of X for each point in the sample
space is
ll

s HHH HHT HTH THH TTH THT HTT TTT

A

X(s) 3 2 2 2 1 1 1 0

The range for the random variable X is X = {0, 1, 2, 3}. Assuming that all eight points
in S have probability 81 , by simply counting in the above display we see that the induced
probability function on X is given by

2
 x 0 1 2 3
1 3 3 1
PX (X = x) 8 8 8 8

ld
The previous illustrations had both a finite S and finite X , and the definition of PX was
straightforward. Such is also the case if X is countable. If X is uncountable, we define the
induced probability function, PX , in a manner similar to (1). For any set A X ,

or
PX (X A) = P ({s S : X(s) A}). (2)

This does define a legitimate probability function for which the Kolmogorov Axioms can be
verified.

Distribution Functions W
TU
Definition of Distribution The cumulative distribution function (cdf) of a random variable
X, denoted by FX (x), is defined by

FX (x) = PX (X x), for all x.

JN

Example 1.5.2 (Tossing three coins) Consider the experiment of tossing three fair coins,
and let X =number of heads observed. The cdf of X is




0 if < x < 0





1
if 0 x < 1

ll

FX (x) = 1 if 1 x < 2

2




A

7

8
if 2 x < 3




1 if 3 x < .

3
 Remark:

1. FX is defined for all values of x, not just those in X = {0, 1, 2, 3}. Thus, for example,
7
FX (2.5) = P (X 2.5) = P (X = 0, 1, 2) = .
8

ld
2. FX has jumps at the values of xi X and the size of the jump at xi is equal to
P (X = xi ).

or
3. FX = 0 for x < 0 since X cannot be negative, and FX (x) = 1 for x 3 since x is
certain to be less than or equal to such a value.

FX is right-continuous, namely, the function is continuous when a point is approached

W
from the right. The property of right-continuity is a consequence of the definition of the cdf.
In contrast, if we had defined FX (x) = PX (X < x), FX would then be left-continuous.

Theorem 1.5.3
The function FX (x) is a cdf if and only of the following three conditions hold:
TU
a. limx F (x) = 0 and limx F (x) = 1.

b. F (x) is a nondecreasing function of x.

c. F (x) is right-continuous; that is, for every number x0 , limxx0 F (x) = F (x0 ).
JN

Example 1.5.4 (Tossing for a head) Suppose we do an experiment that consists of tossing
a coin until a head appears. Let p =probability of a head on any given toss, and define
X =number of tosses required to get a head. Then, for any x = 1, 2, . . .,

P (X = x) = (1 p)x1 p.
ll

The cdf is
x
X x
X
FX (x) = P (X x) = P (X = i) = (1 p)i1 p = 1 (1 p)x .
A

i=1 i=1

It is easy to show that if 0 < p < 1, then FX (x) satisfies the conditions of Theorem 1.5.3.
First,
lim FX (x) = 0
x

4
since FX (x) = 0 for all x < 0, and

lim FX (x) = lim (1 (1 p)x ) = 1,

x x

where x goes through only integer values when this limit is taken. To verify property (b),

ld
we simply note that the sum contains more positive terms as x increases. Finally, to verify
(c), note that, for any x, FX (x + ) = FX (x) if > 0 is sufficiently small. Hence,

lim FX (x + ) = FX (x),

or
0

so FX (x) is right-continuous.

Example 1.5.5 (Continuous cdf)

W
An example of a continuous cdf (logistic distribution) is the function
1
FX (x) = .
1 + ex
It is easy to verify that

lim FX (x) = 0 and lim FX (x) = 1.

TU
x x

Differentiating FX (x) gives

d ex
FX (x) = > 0,
dx (1 + ex )2
showing that FX (x) is increasing. FX is not only right-continuous, but also continuous.
JN

Definition of Continuous Random Variable A random variable X is continuous if FX (x) is a

continuous function of x. A random variable X is discrete if FX (x) is a step function of x.

We close this section with a theorem formally stating that FX completely determines the
probability distribution of a random variable X. This is true if P (X A) is defined only
ll

for events A in B1 , the smallest sigma algebra containing all the intervals of real numbers of
the form (a, b), [a, b), (a, b], and [a, b]. If probabilities are defined for a larger class of events,
A

it is possible for two random variables to have the same distribution function but not the
same probability for every event (see Chung 1974, page 27).

Definition of Identical Random Variables The random variables X and Y are identically dis-
tributed if, for every set A B1 , P (X A) = P (Y A).

5
 Note that two random variables that are identically distributed are not necessarily equal.
That is, the above definition does not say that X = Y .
Example 1.5.9 (identically distributed random variables) Consider the experiment of toss-
ing a fair coin three times. Define the random variables X and Y by

ld
X =number of heads observed and Y =number of tails observed.

For each k = 0, 1, 2, 3, we have P (X = k) = P (Y = k). So X and Y are identically

or
distributed. However, for no sample point do we have X(s) = Y (s).

Theorem 1.5.10 The following two statements are equivalent:

W
a. The random variables X and Y are identically distributed.

b. FX (x) = FY (x) for every x.

Proof: To show equivalence we must show that each statement implies the other. We first
show that (a) (b).
TU
Because X and Y are identically distributed, for any set A B1 , P (X A) = P (Y A).
In particular, for every x, the set (, x] is in B1 , and

FX (x) = P (X (, x]) = P (Y (, x]) = FY (x).

JN

The above argument showed that if the X and Y probabilities agreed in all sets, then
agreed on intervals. To show (b) (a), we must prove if the X and Y probabilities agree on
all intervals, then they agree on all sets. For more details see Chung (1974, section 2.2).
ll
A

dened as:

ld

or

Similar to PMF, probability density function is dened for continuous random

dened as

W
TU

JN

ll

which is dened as

ld

or

TU

JN

ll

We see that the coefcient of



is dened as

ld

or

TU

JN

ll

ld

or

are dened by


TU

JN

ll

ld

or

Dene m.g.f of a r.v. Find m.g.f of (i)

W
TU
JN
ll
A
 ld
or
W
TU

In statistics moment are dened as the mean values of powers o

JN

and is dened as

ll

and is dened as

ld


and is dened as

or

and is dened as

TU

JN

and is dened as

ll

and is dened as

ld

or

TU

JN

Calculate rst three moments about assumed mean 89 for the fo

ll

ld

or

TU

ger. The rst moment about

JN

ll

Compute the rst four central moments for the set 2, 4, 6, 8.:


ld

or

W
TU
From the following frequency distribution, compute rst fo

JN

ll

The rst four moments about mean are:


ld

or
From the data given below compute the rst four moments about

W



















TU

The rst four central moments are:
JN

ll

ld

or

The rst four moments about 4 in a distribution are -1.5,17,-

TU

JN

ll

ld

or

W
Calculate rst four central moments from the following data

TU

JN

ll

ld

or

TU

JN

are rst four moments about origin then rst four moments abo

ll

 ld
Problem 1 The probability density function of a continuous random variable X is given
by f x Ke . Find K and C.D.F of X
x

or
Solution:
Since it is a probability density function,

f x dx 1



x
Ke dx 1

Therefore,

2 Ke x dx 1
0

e x
2K 1
1 0
W
2K 1 K
1
2
.
TU
1
f x ex , x 0
2
1 x
e , 0x
2
x
1 1
For x 0, the C.D.F is F x 2e dx 2 e
x x
JN

0 x
1 1 1 1 1
For x > 0, F x 2e dx 2e
x x
dx (1 e x ) (2 e x )
0
2 2 2

Problem 2 X and Y are independent random variables with variance 2 and 3. Find the
variance of 3 X 4Y .
Solution:
V 3 X 4Y 9Var ( X ) 16Var (Y ) 24Cov( XY )
ll

9 2 16 3 0 ( X & Y are independent cov( XY ) 0 )

18 48 66.
A

Problem 3 A Continuous random variable X has a probability density function

F x 3 x 2 ; 0 x 1. Find a such that P x a P x a
Solution:
We know that the total probability =1
 Given P X a P X a K ( say )
Then K K 1
1
K
2
1 1
i.e., P X a & P X a

ld
2 2
1
Consider P X a
2
a
1
i.e., f x dx

or
0
2
a
1
3x dx 2
2

0
a
x3
1/ 3
1 1 1
3 a3 a .
3 0 2

Problem 4 A random variable X has the p.d.f f x given by f x

Find the value of C and cumulative density function of X .

Solution:

0W
2 2
Cxe x ; if x 0
; if x 0
TU
Since f x dx 1

Cxe
x
dx 1
0

C x e x e x 1

C 1
0
JN

xe x ; x 0
f x
0 ;x 0
x x
C.D.F F x f x dt te t dt te t e t xe x e x 1
x

0
0 0

= 1 1 x e x
for x 0.
1
x 1 ; 1 x 1
Problem 5 If a random variable X has the p.d.f f x 2
ll

, find the
0 ; otherwise
mean and variance of X .
A

Solution:
1 1 1
x x 1 dx x 2 x dx
1 1
Mean= xf x dx
1
2 1 2 1
 1
1 x3 x 2 1

2 3 2 1 3
1
1 x x3
1 1 4
2 x f2
x dx
1
2 1
x 3
x 2
dx
2 4

3 1
1

ld
1 1 1 1 1

2 4 3 4 3
1 2 1
.
2 3 3

or

2
Variance 2 1
1 1 3 1 2
= .
3 9 9 9

Find the moment generating function.

Solution:

W
M X t E etx etx f x dx etx 2e 2 x dx
2e2 x ; x 0
Problem 6 A random variable X has density function given by f x
0 ; x 0
.
TU
0 0

2 e
t 2 x
dx
0

e t 2 x 2
2 ,t 2.
t 2 0 2 t
Problem 7 If X is a Poisson variate such that P X 2 9 P X 4 90 P X 6 , find
JN

the variance.
Solution:
Given P X 2 9 P X 4 90 P X 6
e 2 e 4 e 6
9 90
2! 4! 6!
3 4 0
4 2

2 4 2 1 0
ll

2 1 (or ) 2 4
Variance 1 ( 2 cannot be negative)
A

Problem 8 Comment the following: The mean of a binomial distribution is 3 and

variance is 4
Solution:
In a binomial distribution, mean (np) variance (npq).
 Since variance 4 & mean 3 , we have variance mean. Therefore, the given statement
is wrong.

Problem 9 If X and Y are independent binomial variates

1 1
B 5, and B 7, find P X Y 3
2 2

ld
Solution:
1
X Y is also a binomial variate with parameters n1 n2 12 & p
2
3 9
1 1
P X Y 3 12C3

or
2 2
55
10
2
4
Problem 10 If X is uniformly distributed with Mean 1 and Variance , find P X 0
Solution:

EX
ba
2
ba
and V X
12
W
If X is uniformly distributed over a, b , then
b a
2
3

1 a b 2
TU
2
b a
2
4
b a 16

2

12 3
a b 2 & b a 4 We get b 3, a 1
a 1& b 3 and probability density function of x is
1
JN

; 1 x 3
f x 4
0 ; Otherwise
3
1 1 3 3
P x 0 dx x 0 .
0
4 4 4
3
Problem 11 If X is N 2,3 ind P Y where Y 1 X .
2
ll

Solution:
3 3
P Y P X 1
2 2
A

X 2
P X 2.5 P Z 0.17 , where Z
3
0.5 P 0 Z 0.17
0.5 0.0675 0.4325
 1
Problem 12 If the probability is that a man will hit a target, what is the chance that he
4
will hit the target for the first time in the 7th trial?
Solution:
The required probability is
P FFFFFFS P F P F P F P F P F P F P S

ld
6
3 1
q p . 0.0445 .
6

4 4
Here p probability of hitting target and q 1 p .

or
Problem 13 A random variable X has the following probability function:
Values of
X : 0 1
PX : 0 K
2 3
2K
4 5
2K
6
3K
7
K2

W
2K 2 7K 2 K
Find (i) K , (ii) Evaluate P X 6 , P X 6 and P 0 X 5
(iii). Determine the distribution function of X .
Solution:
(i)
TU
7
Since P x 1,
x 0

K 2 K 2 K 3K K 2 K 2 7 K 2 K 1
2

10K2+9K1=0
1
K or K 1
10
JN

1
As P x cannot be negative K
10
(ii)
P X 6 P X 0 P X 1 ... P X 5
1 2 2 3 1 81
...
10 10 10 10 100 100
Now P X 6 1 P X 6
ll

81 19
1
100 100
Now P 0 X 5 P X 1 P X 2 P X 3 P X 4
A

K 2 K 2 K 3K
8 4
8K .
10 5
(iii) The distribution of X is given by FX x defined by
 FX x P X x
X : 0 1 2 3 4 5 6 7

FX x : 0
1 3 5 4 81 83
1
10 10 10 5 100 100

ld
Problem 14 (i). If the probability distribution of X is given as
X : 1 2 3 4
P X : 0.4 0.3 0.2 0.1
Find P 1/ 2 X 7 / 2 X 1

or
x
; x 1, 2,3, 4,5
(ii). If P x 15
0 ; elsewhere
find (a) P X 1 or 2 and (b) P 1/ 2 X 5 / 2 x 1
Solution:

(i) P 1/ 2 X 7 / 2 / X 1

W
P 1/ 2 X 7 / 2 X 1
P X 1
P X 2 or 3
P X 2,3 or 4
P X 2 P X 3
TU

P X 2 P X 3 P X 4
0.3 0.2 0.5 5
.
0.3 0.2 0.1 0.6 6

(ii) (a) P X 1 or 2 P X 1 P X 2
JN

1 2 3 1


15 15 15 5
1 5
P X X 1
1 2 2
(b) P X / x 1
5
2 2 P X 1
P X 1or 2 X 1

P X 1
ll

P X 2

1 P X 1
A

2 /15 2 /15 2 1
.
1 1/15 14 /15 14 7
 Problem 15 A random variable X has the following probability distribution
X : 2 1 0 1 2 3
P X : 0.1 K 0.2 2 K 0.3 3K
a) Find K , b) Evaluate P X 2 and P 2 X 2
b) Find the cdf of X and d) Evaluate the mean of X .

ld
Solution:
a) Since PX 1
0.1 K 0.2 2 K 0.3 3K 1
6 K 0.6 1

or
6 K 0.4
0.4 1
K
6 15
b) P X 2 P X 2, 1, 0 or 1

10 15 5 15
3 2 6 4 15 1
30

30 2
W
P X 2 P X 1 P X 0 P X 1
1 1 1 2
TU
P 2 X 2 P X 1, 0 or 1
P X 1 P X 0 P X 1
1 1 2

15 5 15
1 3 2 6 2

JN

15 15 5
c) The distribution function of X is given by FX x defined by
FX x = P X x
X : 2 1 0 1 2 3

FX x :
1 1 11 1 4
1
10 6 30 2 5
ll

d) Mean of X is defined by E X xP x

1 1 1 2 3 1
E X 2 1 0 1 2 3
A

10 15 5 15 10 5
1 1 2 3 3 16
.
5 15 15 5 5 15
 Problem 16 X is a continuous random variable with pdf given by
Kx in 0 x 2
2 K in 2 x 4

FX
6 K Kx in 4 x 6
0 elsewhere

ld
Find the value of K and also the cdf FX x .
Solution:

Since F x dx 1

or

2 4 6

Kxdx 2Kdx 6k kx dx 1
0 2 4

x
2 2 6
x2
6

K 2 x 2 6 x 1
4

2 0 4
2 4

K 2 8 4 36 18 24 8 1

x
8K 1
K
1
8
W
We know that FX x f x dx
TU

x
If x 0 , then FX x f x dx 0

x
If x 0, 2 , then FX x f x dx

JN

0 x
FX x f x dx f x dx
0
0 x 0 x
1
0dx Kxdx
0
0dx

8 0
xdx
x
x2 x2
,0 x 2
16 0 16
ll

x
If x 2, 4 , then FX x f x dx

A

0 2 x
FX x f x dx f x dx f x dx
0 2
0 2 x
0dx Kxdx 2 Kdx
0 2
 2
x2 x
2 x x
x 1
dx dx
0
8 2
4 16 0 4 2
1 x 1

4 4 2
x 4 x 1

ld
,2 x4
4 16 4
0 2 4 x
If x 4, 6 , then FX x 0dx Kxdx 2Kdx k 6 x dx
0 2 4

or
2 4 x
x 1 1
dx dx 6 x dx
0
8 2
4 4
8
2 x
x2 x 6x x2
4

16 0 4 2 8 16 4

1
1
4
1 6 x x2
2 8 16

W
3 1

4 16 8 12 x x 2 48 16

16
16
x 12 x 20
,4 x 6
TU
0 2 4 6 x
If x 6 , then FX x 0dx Kxdx 2Kdx k 6 x dx 0dx
0 2 4 6
1, x 6
0 ;x0
2
x ;0 x 2
16
JN

1
FX x x 1 ;2 x4
4
1
16 20 12 x x ; 4 x 6
2

1 ;x 6
ll

K
, x
Problem 17 A random variable X has density function f x 1 x 2
0 , Otherwise
A

Determine K and the distribution function. Evaluate the probability P x 0 .

Solution:

Since f x dx 1


K
1 x

2
dx 1

dx
K 1

1 x2
K tan 1 x

ld
1

K 1
2 2
K 1

or
1
K

x x
K
FX x f x dx 1 x 2
dx

tan 1 x
1

1
x

tan 1 x

1


2
W
1
2 tan x , x
TU

tan 1 x
1 dx 1
P X 0
0 1 x 2
0

1 1 1
tan 0 .
2 2

Ke 3 x , x 0
JN

Problem 18 If X has the probability density function f x find K ,

0 , otherwise
P 0.5 X 1 and the mean of X .

Solution:

Since f x dx 1

ll

Ke
3 x
dx 1
0

e3 x
A

K
K 1 1 K 3
3 0 3
e3 e 1.5
1 1
P 0.5 X 1 f x dx 3 e
3 x
dx 3 e 1.5 e 3
0.5 0.5
3

Mean of X E x xf x dx 3 xe 3 x dx
0 0

e e 3 x
3 x
3 1 1
3x 1
3 9 0 9 3

ld
2 x , 0 x 1
Problem 19 A random variable X has the P.d.f f x
0 , Otherwise
1 1 1 3 1
Find (i) P X (ii) P x (iii) P X / X
2 4 2 4 2

or
Solution:
1/ 2
1
1/ 2 1/ 2
x2 2 1 1
(i) P x
2 f x dx
0 0
2 xdx 2
2 0 8

4
1/ 2
1 1
1/ 21/ 2
x2
(ii) P x f x dx 2 xdx 2
4

2 1/ 4

1
(iii) P X / X
3
1/ 4

1 1 1 1 3
2 .
8 32 4 16 16
3 1
W
2 1/ 4

P X X P X
4 2

3
4
2
TU
4 1 1
P X P X
2 2
1
3
1 1
x2 9 7
P X f x dx 2 xdx 2 1
4 3/ 4 3/ 4 2 3/ 4 16 16
1
1
1 1
x2 1 3
P X f x dx 2 xdx 2 1
2 1/ 2 2 1/ 2 4 4
JN

1/ 2

7
3 1 7 4 7
P X / X 16 .
4 2 3 16 3 12
4
Problem 20 A Man drawn 3 balls from an urn containing 5 white and 7 black balls. He
gets Rs.10 for each white ball and Rs.5 for each black ball. Find his expectation.
ll

Solution:
Let X denotes the amount that he expects to receive
W B
A

5 7
3B 1W & 2 B 2W &1B 3W
X Rs15 Rs 20 Rs 25 Rs30
 7 65
1 2 3
7C3 7
P X 15 P 3 Black balls
12C3 12 1110 44
1 2 3
7C2 .5C1 21
P X 20 P 2 B 1W

ld
12C3 44
7c 2.5c 2 14
P x 25 P 2W 1B
12c3 44
5C3 2
P X 30 P 3W

or
12C3 44
E X xP x
7 21 14 2
15 20 25 30
44 44 44 44
935
Rs.21.25
44

W
Problem 21 From an urn containing 3 red and 2 black balls, a man is to draw 2 balls at
random without replacement, being promised Rs.20/- for each red ball he draws and
Rs.10/- for each black ball. Find his expectation.

Solution:
TU
Let X denotes the amount he receives
R B
3 2
2B 1R 1B 2R
X Rs 20 Rs30 Rs 40
JN

2 1
2C2 1 2 1
P X 20 P 2 Black balls
5C2 5 4 10
1 2
3C 2C1 6
P X 30 P 1 Re d & 1 Black ball 1
5C2 10
3C 3
P X 40 P 2 Re d balls 2
ll

5C2 10
E x xP x
1 6 3
A

= 20 30 40
10 10 10
20 180 120 320

10 10
E x Rs.32 /
 Problem 22 The elementary probability law of a continuous random variable is
f x y0 e , a x , b 0 where a, b and y0 are constants. Find y0 , the rth
b x a

moment about the point x a and also find the mean and variance.

Solution:

ld
Since the total probability is unity,

f x dx 1

b x a
y0 e dx 1

or
0

e b x a
y0 1
b 0
1
y0 1
b
y0 b.

r ( rth moment about the point x a )

b x a e b x a dx
r
W

x a f x dx

r
TU
a

Put x a t , dx dt , when x a, t 0 ; x , t

b t r ebt dt
0

r 1 r!
b r 1

br
JN

b
In particular r 1
1
1
b
2
2 2
b
1
Mean a 1 a
b
ll

2
Variance 2 1
2 1 1
A

2
2 2.
b b b

Problem 23 The first four moments of a distribution about x 4 are 1,4,10 and 45
respectively. Show that the mean is 5, variance is 3, 3 0 and 4 26 .
 Solution:
Given 1 1, 2 4, 3 10, 4 45
r r th moment about to value x 4
Here A 4
Hence mean A 1 4 1 5

ld

2
Variance 2 2 1
4 1 3 .

3
3 3 321 2 1

or
10 3 4 1 2 1 0
3

2 4
4 4 4 3 1 6 2 1 3 1

45 4 10 1 6 4 1 3 1

W
2 4

4 26 .

Problem 24 A continuous random variable X has the p.d.f f x kx 2e x , x 0. Find the

rth moment of X about the origin. Hence find mean and variance of X.
TU
Solution:

Kx e
2 x
Since dx 1
0

e x e x e x
K x2 2x 2 1
1 1 1 0
JN

1
2K 1 K .
2

r x r f x dx
0

1 r2 x
2 0
x e dx

r 2 !
ll

1 x r 3 1

20
e x dx
2
3!
Putting r 1 , 1 3
A

2
4!
r 2 , 2 12
2
Mean = 1 3

2
Variance = 2 1

i.e., 2 12 3 12 9 =3
2

Problem 25 Find the moment generating function of the random variable X, with

ld
x for 0 x 1

probability density function f x 2 x for 1 x 2 . Also find 1 , 2 .
0
otherwise
Solution:

or

M X t e f x dx
tx

1 2
e xdx etx 2 x dx
tx

0 1

xe

t
t t
e e 1 e
e
t 0
2t
2 2 2 2
t t t
et 1
t
tx

t
e e
t t
etx
t
t

2
etx
tx
2 2 x (1) 2
t 1
1

W 2

TU
t
2
t t2 t3
1 ... 1
1! 2! 3!
2
t t2 t3
1 ...
2! 3! 4!
JN

t
1 coeff . of 1
1!
t2 7
2 coeff . of .
2! 6
Problem 26 Find the moment generating function and rth moments for the distribution
whose p.d.f is f x Ke x , 0 x . Find also standard deviation.
ll

Solution:
Total probability=1

ke x dx 1
A

0

e x
k 1 k 1
1 0

M X t E etx etx e x dx e t 1 x dx
0 0

e t 1 x
1
, t 1
t 1 0 1 t

ld
1 t 1 t t 2 ... t r ...
1

t2
r coeff . of r!
r!
When r 1 , 1 1! 1

or
r 2 , 2 2! 2
Variance 2 1 2 1 1
Standard deviation = 1.

Solution:

M X t E e tx

e f x dx e
tx
W
Problem 27 Find the moment generating function for the distribution whose p.d.f is
f x e x , x 0 and hence find its mean and variance.

x tx
e dx
TU
0 0


x t e x t
e dx
0 t 0 t
d 1
Mean 1 M X t 2

dt t 0 t t 0
JN

d2 2 2
2 2 M X t 3
2
dt t 0 t t 0

2
2 1 1
Variance 2 1 2 2 2

1 2x
e ,x 0
Problem 28 Let the random variable X have the p.d.f f x 2
0
ll

, otherwise.
Find the moment generating function, mean & variance of X .
A

Solution:

M X t E etx
1 x/ 2
e f x dx e
tx tx
e dx
0
2

t1 x
1 e 2
t 1 x
1 1 1
e 2 dx , if t .
20 2 1 1 2t 2
t
2
0

ld
d
E X M X t 2
2
dt t 0 1 2t t 0

d2 8
E X 2 2 M X t 3
8
dt t 0 1 2t t 0

or
Var X E X 2 E X 8 4 4 .
2

Problem 29 a) Define Binomial distribution Obtain its m.g.f., mean and variance.
b) Six dice are thrown 729 times. How many times do you expect at least 3
dice show 5 or 6 ?
Solution:

values and its probability mass

P X x nC x p q , x 0,1, 2,..., n and q 1 p.
x n x

M.G.F of Binomial Distribution about origin is

n
W
a) A random variable X is said to follow binomial distribution if it assumes only non-
negative function is given by

M X t E etx etx P X x
TU
x 0
n
nC x x P x q n x etx
x 0
x

nC x pe q
n
t n x

x 0

M X (t ) q pet
n
JN

Mean of Binomial distribution

Mean E X M X 0

n q pet pet np Since q p 1

n 1

t 0
E X 2 M X 0

n n 1 q pet pe npet q pet

n 2 n 1
ll

t 2
t 0
E X 2 n n 1 p 2 np
A

n 2 p 2 np 1 p n 2 p 2 npq
Variance E X 2 E X npq
2

Mean np ; Variance npq

b) Let X : the number of times the dice shown 5 or 6
 1 1 1
P 5 or 6
6 6 3
1 2
P and q
3 3
Here n 6
To evaluate the frequency of X 3

ld
By Binomial theorem,
r 6 r
1 2
P X r 6Cr where r 0,1, 2...6 .
3 3
P X 3 P 3 P 4 P 5 P 6

or
3 3 4 2 5 6
1 2 1 2 1 2 1
6C3 6C4 6C5 6C6
3 3 3 3 3 3 3
0.3196
Expected number of times at least 3 dies to show 5 or 6 N P X 3

variance.
W
Problem 30 a) Find the m.g.f. of the geometric distribution and hence find its mean and

b) Six coins are tossed 6400 times. Using the Poisson distribution, what is the
approximate probability of getting six heads x times?
729 0.3196 233 .
TU

Solution:
a) M.G.F about origin M X t E etx

x e
etx
x 0 x!

e t x
JN

e e e e
t

x 0 x!

M X t e
e 1
t

Mean and variance using M.G.F

Mean E X M X 0

e et
et 1

t 0
ll

E X 2 M X 0 et e et
2 et 1 et 1
e
t 0
A

2
Variance E x 2 E X
2

1
b) Probability of getting one head with one coin .
2
 6
1 1
The probability of getting six heads with six coins
2 64
1
Average number of 6 heads with six coins in 6400 throws np 6400 100
64
Mean of the Poisson distribution 100

ld
By Poisson distribution, the approximate probability of getting six heads x times is given
by
x e 100 e
x 100

P X x , x 0,1, 2,...
x! x!

or
Problem 31 a) A die is cast until 6 appears. What is the probability that it must cast more
than five times?
b) 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 0.8.
(i) What is the probability that the target would be hit on 6th attempt?

Solution: Probability of getting six

1
p & q 1
6
1
6
1
6 W
(ii) What is the probability that it takes him less than 5 shots?

Let x = Number of throws for getting the number 6. By geometric distribution

TU
P X x q x 1 p, x 1, 2,3....
Since 6 can be got either in first, secondthrows.
To find P X 5 1 P X 5
x 1
5
5 1
1 .
x 1 6 6
JN

1 5 1 5 2 1 5 3 1 5 4 1
1
6 6 6 6 6 6 6 6 6
1 5
5

1
6 6 5 5
1 0.4019
1
5 6
6
b) Here p 0.8, q 1 p 0.2
ll

P X r q r 1 p, r 0,1, 2...
(i) The probability that the target would be hit on the 6th attempt P X 6
A

0.2 0.8 0.00026

5

(ii) The probability that it takes him less than 5 shots

P X 5
 4 4
q r 1 p 0.8 0.2
r 1

r 1 r 1

0.8 1 0.2 0.04 0.008 0.9984

Problem 32 a) If X 1 , X 2 are two independent random variables each flowing negative

ld
binomial distribution with parameters r1 , p and r2 , p , show that the sum also follows
negative binomial distribution
b) If a boy is throwing stones at a target, what is the probability that his 10th throw is
his 5th hit, if the probability of hitting the target at any trial is 0.5?

or
Solution:
a) Let X 1 be a negative binomial variate with r1 , p and X 2 be another negative
binomial variate wit r2 , p and let them be independent.
Then M X1 t q pet
r1

M X 2 t q pet
W r2

Then M X1 X 2 t M X1 t M X 2 t
= q pet
r1 r2
, which is the m.g.f of a negative binomial
variable with r1 r2 as parameter. This proves the result.
TU

b) Since the 10th throw should result in the 5th success

(i.e.) the 5th hit, the first 9 throws should have resulted in 4 successes and 5 failures.
1
Hence we have x 5, r 5, p q
2
Required probability P X 5
x r 1 r x
JN

By N.B.D, P X x p q
x
5 5 1 5 5
10
1
p q 9C 4 0.123 .
5 2
Problem 33 a) State and prove the memoryless property of exponential distribution.
b) A component has an exponential time to failure distribution with mean of 10,000
hours.
ll

(i) The component has already been in operation for its mean life. What is the
probability that it will fail by 15,000 hours?
(ii) At 15,000 hours the component is still in operation. What is the probability that it
will operate for another 5000 hours.
A

Solution:
a) Statement: If X is exponentially distributed with parameters , then for any two
positive integers s and t, P x s t / x s P x t
 Proof:
e x , x 0
The p.d.f of X is f x
0 , Otherwise


P X t e x dx e x e t
t

ld
t

P x s t x s
P X s t / x s
P x s
P X s t e s t
s e t
P X s

or
e
Px t
b) Let X denote the time to failure of the component then X has exponential distribution
with Mean 1000 hours.
1 1
10, 000

1

10, 000

The p.d.f. of X is f x 10, 000

0
e

x
10,000
W
,x 0

, otherwise
(i) Probability that the component will fail by 15,000 hours given it has already been in
operation for its mean life P x 15, 000 / x 10, 000
TU

P 10, 000 X 15, 000

P X 10, 000
15,000

f x dx
e1 e 1.5
10,000


e1
JN

f x dx
10,000

0.3679 0.2231
0.3936 .
0.3679
(ii) Probability that the component will operate for another 5000 hours given that
it is in operation 15,000 hours P X 20,000 / X 15, 000
P x 5000 [By memoryless property]
ll

f x dx e
0.5
0.6065
5000
A

Problem 34 The Daily consumption of milk in a city in excess of 20,000 gallons is

approximately distributed as a Gamma variate with parameters 2 and
 1
. The city has a daily stock of 30,000 gallons. What is the probability
10, 000
that the stock is insufficient on a particular day?

Solution:

ld
Let X be the r.v denoting the daily consumption of milk (is gallons) in a city
Then Y X 20, 000 has Gamma distribution with p.d.f.
y
1
f y 2 1
y e 10,000
,y0
10, 000 2
2

or
y

10,000
ye
f y , y 0.
(10, 000) 2
The daily stock of the city is 30,000 gallons; the required probability that the stock is
insufficient on a particular day is given by
P X 30, 000 P Y 10, 000

Put Z
y
10, 000

, then dz

10,000

dy
10, 000
W ye
g y dy 10, 000 dy

10,000

y
10,000

2
TU

P X 30, 000 ze z dz
1
2
ze z e z
1 e

Problem 35 a) suppose that the lifetime of a certain kind of an emergency backup battery
JN

(in hours) is a r.v. X , having the Weibull distribution with parameter 0.1 and 0.5 .
(i) Find the mean life time of these batteries
(ii)The probability that such a battery will last more than 800 hours.
b) Each of the 6 tubes of a radio set has the life length (in years) which may be
considered as a r.v that follows a Weibull distribution with parameter 25 and 2 .If
these tubes function independently of one another, what is the probability that no tube
will have to be replaced during the first 2 months of service?
Solution:
ll

a) For the p.d.f of X in the form f x x 1e x , x 0

1
1

Mean
1
A

1
1
i) Mean life time 0.1

0.5 1
0.5
0.1 3 100 2
2
 200 hours
ii) Probability that a battery will last more than 300 hours P X 300

f x dx x
1 x
e dx
300 300

ld
0.1 0.5 x
0.5 0.1 x
0.5
e dx
300

Put y 0.1 x 0.5

or
Then dy 0.1 0.5 x 0.5 dx

P X 300 e y dy
(0.1)(300)0.5

e 0.177
0.5
0.1 300

b) Let X be life length of the tube

Then the p.d.f of X is given by f x x 1e x , x 0
i.e., f x 50 xe 25 x , x 0
P [a tube is not replaced during the first 2 months]

P X ,
1
2

1
W
since two months yrs
TU
6 6

50 xe25 x dx
2

1
6

25

e 25 x 2
1
e 36

6
JN

P[all 6 tube are not replaced during first 2 months]

e 25 / 36 e 25/ 6 0.0155 .
6

Problem 36 Support that the service life (in hours) of a semi conductor is a random
variable having the Weibull distribution with 0.025 and 0.5 , what is the
probability that such a semi conductor will be in working condition after 4000 hours?
ll

Solution:
Let X be the service life of the conductor.
Then X has Weibull distribution with parameter and and whose density is given by
A

f x x 1e ax

Here =0.025 and =0.5

P [Conductor will be in working after 4000 hours] P x 4000 f x dx
4000
 We know that P X a e a

Required probability e
0.5
0.025 4000

e 1.58 0.2057
Problem 37 a) The amount of time that a camera will run without having to be reset is a
random variable having exponential distribution with 50 days. Find the probability

ld
that such a camera will (i) have to be reset in less than 20 days (ii) not has to be reset in at
least 60 days.
b) Subway trains on a certain line run every half an hour between midnight and six in
the morning. What is the probability that a man entering the station at a random time
during this period will have to wait at least 20 minutes.

or
Solution:
a) Let X be the time that the camera will run without having to be reset.
The X is a random variable with exponentially distributed with
50 days. The p.d.f of X is given by
1 x
e ,x 0
f x
0
x0 W
(i) P[Camera will have to be reset in less than 20 days]
P Camera will run for less than 20 days
P X 20
TU

20
1 50x
2

0 50
e dx 1 e 5
0.3297

(ii) P[Camera will not have to be reset in at least 60 days]

P camera will run for atleast 60days

1 50x
6
P X 60
JN

60 50 e dx e 5
0.3012

b) Let X denote the waiting time in minutes for the next train. Under the assumption
that a man arrives at the station at random time, X is uniformly distributed on (0, 30) with
1
, 0 x 30
p.d.f. f x 30
0 , Otherwise
ll

The probability that he has to wait at least 20 minutes

P X 20
30
1 1
f x dx 30 30 20 3
A

20

Problem 38 If X is a random variable uniformly distribution in (-1,1), find the p.d.f. of

y cos x
Solution:
 1
, 1 x 1
The p.d.f of X is f X x 2
o , otherwise
y cos x
dy
sin x 1 y 2

ld
dx
dx 1

dy 1 y 2
1
Range of Y : 1 x 1 1 cos 1 y 1

or

cos 1 y
1 y 1
1
, 1 y 1
p.d.f of Y is fY y 2 1 y 2

Solution:
0

W
, Otherwise

Problem 39 If X is uniformly distributed in , , find the p.d.f of Y tan X .
2 2
TU
1
, x
a) The p.d.f of X is f X x 2 2
0 , Otherwise
y tan x x tan 1 y

dx 1
JN

dy 1 y 2
dx
p.d.f. of fY y f X x
dy
1 1
, y
1 y2
(This distribution of y is called Cauchy distribution)
2 x, 0 x 1
ll

Problem 40 a) Given the r.v. X with density function f X x .

0 , otherwise
Find the p.d.f of (i) Y 8 X 3 (ii) Y 3 X 1 .
A

b) If the r.v. X follows an exponential distribution with parameter 2, prove that Y X 3

follows Weibull distribution with parameters 2 and 1/3.

Solution:
 2
y 3
2
dy
a) (i) y 8 x
3
8.3 x 24 x 24 6 y 3
2 2

dx 8
2
dy 1 3
y
dx 6

ld
dx 1 23
The p.d.f of Y fY y f X x 2x y
dy 6
1 2
1
y3. y 3
6

or
1 1
fY y y 3 , 0 y 8
6
dx 1 y 1
(ii) y 3 x 1 and x
dy 3 3
dx
p.d.f of Y fY y f X x

2
3
2e2 x , x 0
b) The p.d.f of X is f X x
0 , otherwise
W
dy

y 1 y 1 , 1 y 4
1
3 9
2
TU
dy dx 1 1 23
Then 3x 2
y
dx dy 3x 2 3
dx
The p.d.f of Y is fY y f X x
dy
1 23 2 x
y .2e
JN

3
2 1
1
y 3 2e 2 y 3
3
1 1 3
1 1

fY y 2 y 3 e2 y
3
1
Taking 2 and the p.d.f of y is in the form
3
ll

fY y y e
1 y
. Since x 0 , y 0
1
The r.v. Y X 3 follows Weibull distribution with parameters 2 and .
3
A
 ld
or

W
TU

JN

is xed for all


ll

ld

or


of trial is nite and xed.

TU
If on an average one ship in every ten is wrecked, nd the probability of a







JN

ll

ld

or

W











TU

JN

ll

ld

or

TU














JN

ll

ld




Karl Pearson coefcients,

or

TU

JN

ll

ld

or

TU










JN

ll

ld



mpling from nite

or
population with replacement or sampling from an innite population with or witho

W




TU




ve or six?


JN

1024 investigators each take 10 individuals to nd out if they read newspa






ll

ld

Assuming head as a success in 512 tosses of 8 coins together nd the exp

or

TU

There are 64 beds in a garden and 3 seeds of particular type of ower a

The probability of a ower being blue is 1/4. Find the number of beds with 3,2,1 and 0 blue owers.


JN

Estimate the probability of a mouse being a female and nd the expected freq




ll

In a binomial distribution mean is 3 and variance is 2, nd the remaining cons


test the goodness of t.

ld

or
mean and standard deviation of the tted distribution?

TU

m is some xed number, then the probability of x successes is

JN

ll

 tends to innity, we have

ld






is xed number.

or

TU

JN

ll

ld

or

TU



can nd all the other constants of this distribution.

JN

ll

ld

or

W












TU

JN

ll

To nd the probabilities for rst of all we obtain the value of

A

 After nding the value of or p(0) we nd p(1), p(2), ... etc. By the relation

ld

or

W











TU

JN

Poisson distribution calculate the probabilities of nding a product (i) without







ll

ld

or

TU

JN

ll

ld

or

TU

JN

ll

ld

or

A car hire rm has two cars which it hires out day by day. The number of
TU









JN

ll

ld

or

TU



The computations may be carried out up to innite successes, but it is clea

JN

ll

ld

or

TU











JN

To make the total probability 1, we can nd the probability of 4 and more ins


ll

ld

or

If the proportion of defective items in a bulk is 4 percent nd the probability

W




follow Poisson distribution , nd the probability that there will be three or more

bution nd the probability that during one particular minute exactly 6 customer
TU
After correcting the errors of the rst 50 pages of a book, it is found th






JN

ll






distribution to nd the chance of his making a mistake if (i) less than 1% of the lette

ld
250 passengers have made reservations for a ight from Delhi to Mum

or

TU

JN

when n approaches innitely is known as the normal curve. Thus, the no

n increases to innity.
nd innite binomial distribu-





The normal distribution was rst discovered by the English Mathematician D

ll

ld



is very difcult to compute the expected frequencies. This difculty is ove

or

TU






JN

is equal to the difference between median and rst quartile






ll

7. Points of Inexion : Near the mean value, the normal curve is concav




Based on mean and standard deviation the specic area under normal c

ld

or

TU

JN

ll

ld

dene a new random variable

or

in gure 3:

W
In the table Area under Normal Curve rst column is



of mean and for negative the area is to left to mean. To nd the area from 1.45 to innite subtract


TU

JN

How would you use normal distribution to nd approximately the frequency

ll

ld

or

TU

JN

ll

ld

or

TU

JN

ll

ld

or

W




TU

JN

ll

ld

or

W




TU







JN

of bulbs might be expected to fail in the rst 700 burning hours ? the table o

ll



In a statistical survey of 1000 small business rms in a city , it was found th


the number of rms whose monthly average sales were less than Rs. 600
the number of rms whose monthly average sales were between Rs. 700

ld

or
A sales-tax ofcer has reported that the average sales of the 500 busin

sales in these business are normally distributed, nd:

TU



JN
ll
A


ld

or

TU

JN

ll

ld

or

TU

JN

ll

ld

or

TU
(, )
Surface ()

Shaded
rectangle
JN

ll

ld

or

TU

JN

ll

(0, 1)

(, 1 )

ld
(1, 0)

or

W







TU

JN

1
ll

Shaded region
A

.5
.5

0
0 .5 1

ld

or

TU

JN

ll

ld

or

TU





JN

ll

ld

or

TU

JN

ll

ld

or

TU






JN

ll

 ld
or
Problem 1 Let X and Y have joint density function f x, y 2, 0 x y 1 .Find the marginal
density functions and the conditional density function Y given X x .

Solution:
Marginal density function of X is given by

fX x f x

f x, y dy

1 1
W
f x, y dy 2dy 2 y x
1
TU
x x

2 1 x , 0 x 1.
Marginal density function of Y is given by

fY y f y f x, y dx

y
JN

2dx 2 y, 0 y 1 .
0

Conditional distribution function of Y given X x is f y x f f x ,xy 2 12 x 1 1 x .

Problem 2 Two random variables X and Y have the joint p.d.f f x, y Ae (2 x y ) , x, y 0 .
Find A.
ll

Solution:
Since f x, y is a joint density function

A

f x, y dxdy 1 .


Ae (2 x y ) dxdy 1
0 0

e2 x
A e y
dy 1
0 2 0

1
A e y dy 1
0
2

ld

1 e y
A 1
2 1 0
1
A 1 1 A 2
2

or
Problem 3 Verify whether X and Y are independent if f x, y kxy, 0 x y, 0 y 4
Solution:
Since f x, y is a joint density function

f x, y dxdy 1 .

4 y
kxydxdy 1
0 0
4
x2
k y dy 1
0 2 0
y
W
TU
4
1 3 1 1
k y dy 1 k y 4 1 k 32 1 k
4

0
2 8 0 32
Marginal density function of X is given by

fX x f x f x, y dy

4
JN

1 y2
4
1 x
xydy x , 0 x y
0
32 32 2 0 4
Marginal density function of Y is given by

fY y f y f x, y dx

y
x2
y
1 1 y3
xydx y , 0 y 4
2 0 64
ll

0
32 32
x y3
f x . f y . f x, y
4 64
A

X and Y are not independent.

Problem 4 Two random variables X and Y have the joint p.d.f f x, y x y,

0 x 1, 0 y 1. Determine the marginal distributions of X and Y.
Solution:
Marginal density function of X is given by

fX x f x f x, y dy

1
1
y2 1
( x y )dy xy x , 0 x 1
2 y 0

ld
0
2
Marginal density function of Y is given by

fY y f y f x, y dx

or
1
1
x2 1
( x y )dx xy y, 0 y 1
0 2 x 0 2
Problem 5 Suppose that the joint density function of X and Y is
Ae x y , 0 x y, 0 y
f x, y Determine A .

Solution:
0 , otherwise

Since f x, y is a joint density function

f x, y dxdy 1 .

y
W
TU
Ae x e y dxdy 1
0 0
y
e x
A e y
dy 1
0 1 0

A e y e2 y dy 1
JN

0

e y e 2 y
A 1
1 2 0
1
A 1 A 2
2

Problem 6 Examine whether the variables X and Y are independent, whose joint density
ll

function is f x, y xe , 0 x, y .
x y 1

Solution:
The marginal probability function of X is
A

f X x f x f x, y dy xe
x y 1
dy
0

e x y 1
0 e e ,
x x
x
x 0
The marginal probability function of Y is

fY y f y f x, y dx x e
x y 1
dx

ld
0

e x y 1 e x y 1
x 2
y 1 0 y 1 0

or
1

y 1
2

1
Here f x . f y e x f x, y
1 y
2

X and Y are not independent.

Solution:
Since y x , x y 2
W
Problem 7 If X has an exponential distribution with parameter 1. Find the pdf of y x

Since X has an exponential distribution with parameter 1, the pdf of X is given by

f X x e x , x 0 f x e x , 1
TU

dx
fY y f X x
dy
e x 2 y 2 ye y
2

fY y 2 ye y , y 0
2
JN

Problem 8 If X is uniformly distributed random variable in , , Find the probability
2 2
density function of Y tanX .
Solution:
Given Y tanX x tan 1 y
dx 1

dy 1 y 2
ll

Since X is uniformly distribution in , ,
2 2
A

1 1
fX x
ba
2 2
 1
fX x , x
2 2
dx 1 1
Now fY y f X x , y
dy 1 y 2
1
fY y , y

ld
1 y 2

Problem 9 If the Joint probability density function of x, y is given by f x, y 24 y 1 x ,

0 y x 1 Find E XY .

or
Solution:
1 1
E xy xyf x, y dxdy y
0 y

1
1 1
24 xy 2 1 x dxdy
0 y

1 y 2 y3
24 y 2 dy .
0 6 2 3
4
15
W x y

Problem 10 If X and Y are random Variables, Prove that Cov X , Y E XY E X E Y

TU

Solution:
cov X , Y E X E X Y E Y
E XY XY YX XY
E XY XE Y YE X XY
JN

E XY XY XY XY
E XY E X E Y E X X , E Y Y

Problem 11 If X and Y are independent random variables prove that cov x, y 0

Solution:
cov x, y E xy E x E y
But if X and Y are independent then E xy E x E y
ll

cov x, y E x E y E x E y Therefore cov x, y 0.

A

Problem 12 Write any two properties of regression coefficients.

Solution:
1. Correction coefficients is the geometric mean of regression coefficients
2. If one of the regression coefficients is greater than unity then the other should be less
than 1.
 y
bxy r and byx r x
x y
If bxy 1 then byx 1

Problem 13 Write the angle between the regression lines.

ld
Solution:
The slopes of the regression lines are
y 1y
m1 r , m2
x r x

or
If is the angle between the lines, Then
x y 1 r 2
tan 2
x y2 r

When r 0 , that is when there is no correlation between x and y, tan (or)
and so the regression lines are perpendicular

W 2

When r 1 or r 1 , that is when there is a perfect correlation ve or ve , 0 and so the

lines coincide.

Problem 14 State central limit theorem

Solution:
If X 1 , X 2 .... X n is a sequence of independent random variable E X i i and
TU

Var X i i2 , i 1, 2,....n and if S n X 1 X 2 ...... X n then under several conditions S n

n n
follows a normal distribution with mean i and variance 2 i2 as n .
i 1 i 1

Problem 15 Fill up the blanks:

JN

i). Two random variables are said to be orthogonal if correction is zero

ii). If X Y then correlation coefficients between them is _____1

Problem 16 The joint probability density function of a bivariate random variable X , Y is

k x y , 0 x 2, 0 y 2
f XY x, y where ' k ' is a constant.
0 , otherwise
ll

i. Find k .
ii. Find the marginal density function of X and Y .
iii. Are X and Y independent?
A

iv. Find fY y X x
and f X x .
Y y
Solution:
(i). Given the joint probability density function of a brivate random variable X , Y is
 K x y , 0 x 2, 0 y 2
f XY x, y
0 , otherwise

Here f XY x, y dxdy 1 K x y dxdy 1

ld
2
2 2
x2
2

0 0 K x y dxdy 1 K 0 2 xy dy 1
0
2
K 2 2 y dy 1

or
0
2
K 2 y y 2 1
0

K 8 0 1
1
K
(ii). The marginal p.d.f of X is given by

1
1
2
f X x f x, y dy x y dy
80

xy
8
y2 1 x
2 0 4
8

2
W
TU
The marginal p.d.f of X is
x 1
, 0 x2
fX x 4
0 , otherwise
The marginal p.d.f of Y is
2
1
fY y f x, y dx x y dx
JN

80
2
1 x2
yx
8 2 0
1 y 1
2 2 y
8 4
The marginal p.d.f of Y is
ll

y 1
, 0 y2
fY y 4
0 , otherwise
A

(iii). To check whether X and Y are independent or not.

f X x fY y
x 1 y 1 f x, y
XY
4 4
Hence X and Y are not independent.
(iv). Conditional p.d.f fY
X
y x is given by
1
x y 1 x y
fY y
X

x
f x ,
fX x
y 8
1

x 1 2 x 1

ld
4

X

fY y
x
1 x y
2 x 1
, 0 x 2, 0 y 2

1 2
0 y 2
x 1 Y X x 1
y

or
(v) P f dy
0

1

1 1 y
2
5

20 2
dy
32
.

1
6 x y , 0 x 2, 2 y 4
f x, y 8
0 , otherwise
W
Problem 17 a) If X and Y are two random variables having joint probability density function

Find (i) P X 1 Y 3

(ii) P X Y 3 (iii) P X 1
TU
.
Y 3
b). Three balls are drawn at random without replacement from a box containing 2 white, 3 red
and 4 black balls. If X denotes the number of white balls drawn and Y denotes the number of
red balls drawn find the joint probability distribution of X , Y .
Solution:
a).
y 3
JN

x 1
P X 1 Y 3 f x, y dxdy
y x
y 3 x 1
1
8 6 x y dxdy
y 2 x 0
3 1
1
6 x y dxdy
820
ll

1
1
3
x2

82 6 x
2
xy dy
0
A

3
1 11 1 11y y 2
3
y dy
822 8 2 2 2
3
P X 1 Y 3
8
 1 3 x
1
(ii). P X Y 3 8 6 x y dydx
0 2
3 x
1 y2
1
6 y xy dx
80 2 2

ld
1 3 x
1 2

6 3 x x 3 x 12 2 x 2 dx
8 0 2

1
1
18 6 x 3 x x 2 9 x2 6 x
10 2 x dx

or
80 2

1
1
9 x2 6 x
18 9 x x 2 10 2 x dx
80 2 2 2
1 7 x2
1

8 0 2
4 x dx

8 2

8 6
2
2
1 7 x 4 x 2 x3 1 7
2
6 0 8 2
1 21 12 1 1 10 5
8 6 24 .
1
6
1

W
P x 1 y 3
TU

(iii). P X 1
Y 3
P y 3
2
The Marginal density function of Y is fY y f x, y dx
0
2
1
6 x y dx
JN

0
8
2
1 x2
6 x yx
8 2 0
1
12 2 2 y
8
5 y
, 2 y 4.
4
ll

x 1 y 3
1
8 6 x y dxdy

P X 1
Y 3 x 0 y 2
A

y 3

fY y dy
y 2
 3 3
3 8 8

3
5 y 1 y2
2 4 dy
4
5y
2 2
3 8 3

ld
.
8 5 5

b). Let X takes 0, 1, 2 and Y takes 0, 1, 2 and 3.

P X 0, Y 0 P ( drawing 3 balls none of which is white or red)

or
P ( all the 3 balls drawn are black)
4C 4 3 2 1 1
3 .
9C3 98 7 21
P X 0, Y 1 P( drawing 1 red ball and 2 black balls)
3C1 4C2 3
=

9C3

14

3C2 4C1 3 2 4 3 1
9C3

98 7
.
7
W
P X 0, Y 2 P( drawing 2 red balls and 1 black ball)

P X 0, Y 3 P( all the three balls drawn are red and no white ball)
TU
3C3 1
=
9C3 84
P X 1, Y 0 P( drawing 1White and no red ball)
2 43
2C 4C2
= 1 1 2
98 7
JN

9C3
1 2 3
12 1 2 3 1
.
98 7 7
P X 1, Y 1 P( drawing 1White and 1 red ball)
23
2C 3C1 9 8 7 2
1
1 2 3 7
ll

9C3
P X 1, Y 2 P( drawing 1White and 2 red ball)
2C1 3C2 2 3 2 1

A

9C3 9 8 7 14
1 2 3
P X 1, Y 3 0 (Since only three balls are drawn)
P X 2, Y 0 P( drawing 2 white balls and no red balls)
 2C2 4C1 1

9C3 21
P X 2, Y 1 P( drawing 2 white balls and no red balls)
2C2 3C1 1

9C3 28

ld
P X 2, Y 2 0
P X 2, Y 3 0
The joint probability distribution of X , Y may be represented as

or
Y 0 1 2 3
X
1 3 1 1
0
21 14 7 84
1 2 1

W
1 0
7 7 14
1 1
2 0 0
21 28

Problem 18 a) Two fair dice are tossed simultaneously. Let X denotes the number on the first
die and Y denotes the number on the second die. Find the following probabilities.

(i) P X Y 8 , (ii) P X Y 8 , (iii) P X Y and (iv) P X Y 6
TU
.
Y 4
b) The joint probability mass function of a bivariate discrete random variable X , Y in given
by the table.
X
Y 1 2 3
1 0.1 0.1 0.2
JN

2 0.2 0.3 0.1

Find
i. The marginal probability mass function of X and Y .
ii. The conditional distribution of X given Y 1 .
iii. P X Y 4
Solution:
a). Two fair dice are thrown simultaneously
ll

1,11, 2 ... 1, 6

2,1 2, 2 ... 2, 6
S , n( S ) 36
A

. . ... .
6,1 6, 2 ... 6, 6

Let X denotes the number on the first die and Y denotes the number on the second die.
 1
Joint probability density function of X , Y is P X x, Y y for
36
x 1, 2,3, 4,5, 6 and y 1, 2,3, 4,5, 6
(i) X Y the events that the no is equal to 8
2, 6 , 3,5 , 4, 4 , 5,3 , 6, 2

ld
P X Y 8 P X 2, Y 6 P X 3, Y 5 P X 4, Y 4
P X 5, Y 3 P X 6, Y 2
1 1 1 1 1 5

or
36 36 36 36 36 36
(ii) P X Y 8
2, 6

3,5 , 3, 6

X Y 4, 4 , 4,5 , 4, 6

5,3 , 5, 4 5,5 , 5, 6



6, 2 , 6,3 , 6, 4 , 6,5 6, 6
W
P X Y 8 P X Y 8 P X Y 9 P X Y 10
P X Y 11 P X Y 12
TU
5 4 3 2 1 15 5

36 36 36 36 36 36 12
(iii) P X Y
P X Y P X 1, Y 1 P X 2, Y 2 ...... P X 6, Y 6
1 1 1 6 1

..........
JN

36 36 36 36 6
P X Y 6 Y 4

(iv) P X Y 6
Y 4

P Y 4
1
Now P X Y 6 Y 4
36
6
P Y 4
36
ll

1

P X Y 6
Y 4
1
36 .
6 6
A

36
b). The joint probability mass function of X , Y is
X 1 2 3 Total
Y
1 0.1 0.1 0.2 0.4
2 0.2 0.3 0.1 0.6

ld
Total 0.3 0.4 0.3 1
From the definition of marginal probability function
PX xi PXY xi , y j
yj

When X 1 ,

or
PX xi PXY 1,1 PXY 1, 2
0.1 0.2 0.3
When X 2 ,
PX x 2 PXY 2,1 PXY 2, 2
0.2 0.3 0.4
When X 3 ,
PX x 3 PXY 3,1 PXY 3, 2
0.2 0.1 0.3
The marginal probability mass function of X is
0.3 when x 1

W
PX x 0.4 when x 2
TU

0.3 when x 3

The marginal probability mass function of Y is given by PY y j PXY xi , y j
xi
3
When Y 1 , PY y 1 PXY xi ,1
xi 1
JN

PXY 1,1 PXY 2,1 PXY 3,1

0.1 0.1 0.2 0.4
3
When Y 2 , PY y 2 PXY xi , 2
xi 1

PXY 1, 2 PXY 2, 2 PXY 3, 2

0.2 0.3 0.1 0.6
Marginal probability mass function of Y is
ll

0.4 when y 1
PY y
0.6 when y 2
A

(ii) The conditional distribution of X given Y 1 is given by

P X x Y 1

P X x
Y 1
P Y 1
From the probability mass function of Y , P y 1 Py 1 0.4
 P X 1 Y 1
When X 1 , P X 1 Y 1
P Y 1
PXY 1,1 0.1
0.25
PY 1 0.4
P 2,1 0.1

ld
When X 2 , P X 2 XY
0.25
Y 1 P 1 0.4
Y

P 3,1 0.2

When X 3 , P X 3
Y 1
P 1
XY

Y

0.4
0.5

or
(iii). P X Y 4 P x, y / x y 4 Where x 1, 2,3; y 1, 2
P 1,1 , 1, 2 , 2,1
PXY 1,1 PXY 1, 2 PXY 2,1
0.1 0.1 0.2 0.4

b) The joint probability density function of

W
Problem 19 a) If X and Y are two random variables having the joint density function
1
f x, y x 2 y where x and y can assume only integer values 0, 1 and 2, find the
27
conditional distribution of Y for X x .
X , Y is given by
TU
2 x2
xy , 0 x 2, 0 y 1
f XY x, y 8 . Find (i) P X 1 , (ii) PX Y and
0
, otherwise
(iii) P X Y 1
Solution:
a). Given X and Y are two random variables having the joint density function
JN

1
f x, y x 2 y (1)
27
Where x 0,1, 2 and y 0,1, 2
Then the joint probability distribution X and Y becomes as follows

Y 0 1 2 f1 x
X
ll

1 2 3
0 0
27 27 27
A

2 3 4 9
1
27 27 27 27
4 5 6 15
2
27 27 27 27
The marginal probability distribution of X is given by f1 X P x, y and is calculated in
j

the above column of above table.

The conditional distribution of Y for X is given by f1 Y y
X x
f x, y
f x
1
and is obtained in

ld
the following table.
X
0 1 2
Y
1 2
0 0

or
3 3
1 3 5
1
9 9 9
1 1 1
2
6 3 2

P Y 0

P Y 1
X 0

X 0

P
P X 0, Y 0

X
P X 0

0,
P X 0
Y 1

0
6
27
2
27
6 3
W
0

1
TU
27
4
P X 0, Y 2 27 2

P Y 2
X 0
P X 0

6

3
27
1
P X 1, Y 0 27 1
0
JN

P Y
X 1 P X 1 9 9
27
3


P Y 1
X 1
P X 1,
P X 1
Y 1
27
9 9 3
3 1

27
5
27 5
ll

P Y 2
X 1
P X 1, Y
P X 1
2
9 9
27
A

2


P Y 0
X 2

P
X 2, Y
P X 2
0
27
12 6
1

27
 4
P X 2, Y 1

P Y 1
X 2

P X 2
27
12 3
1

27
6
27 1

ld
P X 2, Y 2
P Y 2
X 2 P X 2 12 2
27

x2
b). Given the joint probability density function of X , Y is f XY x y xy 2

or
,
8
0 x 2, 0 y 1

(i). P X 1 f X x dx
1

W
1
The Marginal density function of X is f X x f x, y dy
0

1
x 2
f X x xy 2 dy y y 1
0
8
1
xy 2 x 2 y x x2
, 1 x 2
8 0 3 8
TU
3
2
x x2
P X 1 dx x
1
3 8
2
x 2 x3 19
. y x 1 x2
6 24 1 24
JN

(ii) P X Y f XY x, y dxdy y 1
R2

2 x2
1 y

P X Y xy 8 dxdy yx
y 0 x 0
y
x 2 y 2 x3
1
dy x0 x
0
2 24 0
ll

1
1
y4 y3 y5 y4
dy
0
2 24 10 96 0
1 1 96 10 53
A

10 96 960 480
(iii) P X Y 1 f XY x, y dxdy
R3

Where R3 is the region

 1 1 y
2 x2
P X Y 1 y 0 x0 xy 8 dxdy y

1 y
1
x 2 y 2 x 3
dy y 1
y 0
2 24 0

ld
1 1 y 2 y 2 1 y 3
dy x y 1

y 0
2 24

1 1 y2 2 y y2

1 y
3

dy

or
x
0
2 24

1
y 3 y 5 2 y 2 1 1 y 4
x 1
3 5 4 2 96
0

F x, y
0
1 1 1 1

6 10 4 96 480

13
.

W
Problem 20 a) If the joint distribution functions of X and Y is given by
1 e x 1 e y , x 0, y 0
, otherwise
TU
i. Find the marginal density of X and Y .
ii. Are X and Y independent.
iii. P 1 X 3, 1 Y 2 .
b) The joint probability distribution of X and Y is given by
6 x y

f x, y 8
, 0 x 2, 2 y 4

. Find P 1 Y 3
X 2
.
JN

0 , otherwise
Solution:
a). Given F x, y 1 e x 1 e y
1 e x e y e
x y

The joint probability density function is given by

2 F x, y
f x, y
xy
ll

2
1 e x e y e x y
xy
A

y x y
e e
x
x y
e , x 0, y 0
f x, y
0 , otherwise
(ii) The marginal probability function of X is given by
f x fX x

f x, y dy e
x y
dy
0

e x y

ld

1 0

e
x y
0

or
x
e , x0
The marginal probability function of Y is
f y fY y

f x, y dx
e

0
x y

e y , y 0
dx e
x y

f x f y e x e y e f x, y
x y

0
W
X and Y are independent.
TU
(iii) P 1 X 3,1 Y 2 P 1 X 3 P 1 Y 2 [Since X and Y are independent]
3 2
f x dx f y dy.
1 1
3 2
e x dx e y dy
1 1
JN

3 2
e e y
x

1 1 1 1

e 3 e 1 e 2 e 1
5 4 3 2
e e e e

f y x 2 dy
3

b). P 1 Y 3
X 2
ll

1
4
f X x f x, y dy
2
A

6 x y
4
dy
2 8
4
1 y2
6 y xy
8 2 2
 1
16 4 x 10 2 x

8
6 x y

f y
x
f 8 6 x y
x ,
f x
y
6 2x 6 2x

ld
8

dy
3

P 1 Y 3 f y
X 2 x2
1

4 y
3
dy

or
2 2
3
1 y2
4 y
2 2 2
3
1 y2 1 17 11
4 y 14 .
2 2 2 2 2 4

W
Problem 21 a) Two random variables X and Y have the following joint probability density
2 x y, 0 x 1, 0 y 1
function f x, y
0 , otherwise
. Find the marginal probability density function

of X and Y . Also find the covariance between X and Y .

TU
6 x y
b) If f x, y , 0 x 2, 2 y 4 for a bivariate X , Y , find the correlation
8
coefficient
Solution:
2 x y, 0 x 1, 0 y 1
a) Given the joint probability density function f x, y
0 , otherwise
JN

Marginal density function of X is f X x f x, y dy

1
2 x y dy
0
1
y2
2 y xy
2 0
ll

1
2 x
2
3
A

x, 0 x 1
fX x 2
0 , otherwise
 1
Marginal density function of Y is fY y 2 x y dx
0
1
x2
2 x xy
2 0

ld
3
y
2
3
y, 0 y 1
fY y 2
0

or
, otherwise
Covariance of X , Y Cov X , Y E XY E X E Y
1
1 1
3 3 x 2 x3 5
E X xf X x dx x x dx
0 2 2 3 0 12
0
2

0
1

0
1
3

Cov X , Y E XY E X E Y
1 1
E XY xy f x, y dxdy
0 0

E Y yfY y dy y y dy
2
5
12
W
TU
1 1
xy 2 x y dxdy
0 0
1 1

2xy x 2 y xy 2 dxdy
0 0
1
1
2 x 2 y x3 x2 2
y y dy
JN

0 2 3 2 0
1
1 y2
y dy
0
3 2
1
y 2 y y3 1

2 3 6 0 6
1 5 5
Cov X , Y
ll

6 12 12
1 25 1
.
6 144 144
A

E XY E X E Y
b). Correlation coefficient XY
XY
Marginal density function of X is

6 x y 6 2x
4
fX x f x, y dy 2 8 dy 8
Marginal density function of Y is

6 x y 10 2 y
2
fY y f x, y dx dx
0
8 8

ld

6 2x
2 2
Then E X xf X x dx x dx
0
0
8
2
1 6 x 2 2 x3

or
8 2 3 0
1 16 1 20 5
12
8 13 8 3 6
4
10 2 y 1 10 y 2 2 y 3
4
17
E Y y dy

E X

E Y
2

2
2

0
8
2
2

10 2 y
4
y
8
2
dy
8 2

8 3
8
3 2

6 2x
x f x x dx x

2
dx

1 10 y 3 2 y 4

4 2 3
6
1 6 x3 2 x4
8


25
2

3

4
1
0
W 4
2
TU
2
2
5 11
Var X E X E X 1
2 2 2
X
6 36
2
25 17 11
Var Y E Y E Y
2
2
Y
2

3 6 36
6 x y
4 2
E XY xy dxdy
JN

2 0 8
2
1 6 x2 y x3 y x2 y 2
4
dy
82 2 3 2 0
4
1 1 12 y 2 8 y 2 2 y 3
4
8
12 y y 2 y 2 dy
8 2 3 8 2 3 2 3 2
1 64 128 16 16 1 56
ll

96 24
8 3 3 3 3 8 3
7
E XY
A

3
7 5 17

E XY E X E Y 3 6 6

XY
XY 11 11
6 6
 1
XY .
11

Problem 22 a) Let the random variables X and Y have pdf

1
f x, y , x, y 0, 0 , 1,1 , 2, 0 . Compute the correlation coefficient.

ld
3
b) Let X 1 and X 2 be two independent random variables with means 5 and 10 and standard
devotions 2 and 3 respectively. Obtain the correlation coefficient of UV where U 3 X 1 4 X 2
and V 3 X 1 X 2 .

or
Solution:
a). The probability distribution is

X 0 1 2 P Y
Y

0
1
3

0
0
1
3

0
W 0

0
1
1
3
1
3
1
3 3
TU
P X 1 1 1
3 3 3

1 1 1
E X xi pi xi 0 1 2 1
3 3 3
JN

1 1 1 1
E Y yi p j y j 0 1 0
j 3 3 3 3
1 1 1 5

E X 2 xi p xi 0 1 4
2

3 3 3 3
i

5
Var ( X ) E X 2 E X 1
2
2

3 3
1 1 1 1
ll

E Y 2 y j p y j 0 1 0
2

3 3 3 3
j

V Y E Y 2 E Y
1 1 2
2
A

3 9 9
E XY E X E Y
Correlation coefficient XY
V X V Y
E XY xi y j p xi , y j
i j

1 1 1 1
0.0. 0.1.0 1.0.0 1.1. 1.2.0 0.0.0 0.1.0 0.2.
3 3 3 3
1 1
1

ld
XY
3 3 0
2 2

3 9
Correlation coefficient 0 .

or
b). Given E X 1 5, E X 2 10
V X 1 4, V X 2 9
Since X and Y are independent E XY E X E Y
E UV E U E V
Correlation coefficient
Var U Var V
E U E 3 X 1 4 X 2 3E X 1 4 E X 2
3 5 4 10 15 40 55.
E V E 3 X 1 X 2 3E X 1 E X 2
W
3 5 10 15 10 5
TU

E UV E 3 X 1 4 X 2 3 X 1 X 2
E 9 X 1 3 X 1 X 2 12 X 1 X 2 4 X 2
2 2

9 E X 1 3E X 1 X 2 12 E X 1 X 2 4 E X 2
2 2

9E X 9E X X 4E X
2 2
JN

1 1 2 2

9E X 9E X E X 4E X
2 2
1 1 2 2

9 E X 450 4 E X
2 2
1 2

V X E X E X
2 2
1 1 1

E X V X E X 4 25 29
2 2
1 1 1
ll

E X V X E X 9 100 109
2 2
2 2 2

E UV 9 29 450 4 109
A

261 450 436 275

Cov(U , V ) E UV E U E V
275 5 55 0
Since Cov U , V 0, Correlation coefficient 0 .
Problem 23 a) Let the random variable X has the marginal density function
1 1
f x 1, x and let the conditional density of Y be
2 2
1
1, x y x 1, x 0

ld
2
f y . Prove that the variables X and Y are uncorrelated.
x
1, x y 1 x, 0 x 1
2
b) Given f x, y xe
x y 1
, x 0, y 0 . Find the regression curve of Y on X .

or
Solution:
1 1 1
2
x2 2 2
a). We have E X xf x dx xdx 0

1

1 2 1
2 2 2
1

E XY

0 x 1

xydxdy xydxdy
1 x
2

x 1
0

1 x


2

0
1 x
x
2 1 x

0 x

x ydy dx x ydy dx

1 W
TU
2
1
0 2
1 1

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

2
1
0
1 2 x3 x 2 1 x 2 2 x3 2
0
2 3 2 1 2 2 3 0
JN

Since Cov X , Y E XY E X E Y 0 , the variables X and Y are uncorrelated.

b). Regression curve of Y on X is E y x

ll

x yf y x dy

E y

f x, y
A

f y / x
fX X

Marginal density function f X x f x, y dy
0

x y 1
x e dy
0

e x y 1
x e x , x 0
x 0

ld
Conditional pdf of Y on X is f y
x

f x, y xe xy x
fX x
x xe xy
e
The regression curve of Y on X is given by

x

E y yxe xy dy

or
0

e xy e xy
x y 2
x x 0

1 1
E y y and hence xy 1 .
x x

and X on Y .
x

x y
Problem 24 a) Given f x, y 3
0
, 0 x 1, 0 y 2

, otherwise
W
, obtain the regression of Y on X
TU
b) Distinguish between correlation and regression Analysis
Solution:
Regression of Y on X is E Y
X
X yf y x dy

E Y

X ff x,xy
JN

f Y
X

fX x f x, y dy

2
x y 1 y2
2

3 2 0
dy xy
0
3
2 x 1
ll

3
f x, y

f Y
x y
A

X f X x 2( x 1)
y x y
X
2
Regression of Y on X E Y dy
0
2 x 1
 2
1 xy 2 y 3

2 x 1 2 3 0
1 8 3x 4
2x
2 x 1 3 3 x 1

Y xf x y dx

ld

E X

y ff x,yy
f x
Y

or

fY y f x, y dx

1
x y 1 x2
1
dx xy
0
3 3 2 0

f x
y
1 1
y
3 2


2 x y
2 y 1

Regression of X on Y E X
W
Y 2xyy1 dx
1
TU
0
1
1 x2
xy
2 y 1 2 0
1
y
1
2 .
2 y 1 2
JN

b).
1. Correlation means relationship between two variables and Regression is a Mathematical
Measure of expressing the average relationship between the two variables.
2. Correlation need not imply cause and effect relationship between the variables. Regression
analysis clearly indicates the cause and effect relationship between Variables.
3. Correlation coefficient is symmetric i.e. rxy ryx where regression coefficient is not symmetric
4. Correlation coefficient is the measure of the direction and degree of linear relationship
ll

between two variables. In regression using the relationship between two variables we can predict
the dependent variable value for any given independent variable value.

Problem 25 a) X any Y are two random variables with variances x2 and y2 respectively and
A

y x
r is the coefficient of correlation between them. If U X KY and V X , find the
y
value of k so that U and V are uncorrelated.
b) Find the regression lines:
 X 6 8 10 18 20 23
Y 40 36 20 14 10 2
Solution:
Given U X KY
E U E X KE Y

ld
X
VX Y
Y

E V E X X E Y
Y

or
If U and V are uncorrelated, Cov U ,V 0
E U E U V E V 0

E X KY E X KE Y X X Y E X X E Y 0
Y Y

V X
X

Y

W
Cov X , Y KCov X , Y K X V Y 0

Y

E X E X K Y E Y X E X X Y E Y 0

2
E X E X X X E X Y E Y K Y E Y X E X K X Y E Y 0
2

Y Y
TU

K Cov X , Y X V Y V X X Cov x, y
Y Y

V X X r X Y
Y X2 r X2
K
r X Y X Y
r X Y X V Y
JN

Y
X2 1 r
X .
X Y 1 r Y

b).
ll

X Y X2 Y2 XY
6 40 36 1600 240
8 36 64 1296 288
A

10 20 100 400 200

18 14 324 196 252
20 10 400 100 200
 23 2 529 4 46

X 85 Y 122 X 2
1453 Y 2
3596 XY 1226

x 85 y 122
X 14.17 , Y 20.33

ld
n 6 n 6
x2 x
2 2
1453 85
x 6.44
n n 6 6
y2 y
2 2
3596 122

or
y 13.63
n n 6 6
xy 1226
xy 14.17 20.33
r n 6 0.95
x y 6.44 13.63
bxy r

byx r
y
x
x
y
0.95

0.95
6.44
13.63
13.63
6.44
0.45

2.01

The regression line X on Y is

W

x x bxy y y x 14.17 0.45 y y
TU

x 0.45 y 23.32
The regression line Y on X is

y y byx x x y 20.33 2.01 x 14.17
y 2.01x 48.81
JN

Problem 26 a) Using the given information given below compute x , y and r . Also compute
y when x 2, 2 x 3 y 8 and 4 x y 10 .
b) The joint pdf of X and Y is
X
Y -1 1

1 3
ll

0 8 8
2 2
1 8 8
A

Find the correlation coefficient of X and Y .

Solution:
a). When the regression equation are Known the arithmetic means are computed by solving the
equation.
2 x 3 y 8 ------------ (1)
 4 x y 10 ------------ (2)
(1) 2 4 x 6 y 16 ------- (3)
2 3 5 y 6
6
y
5

ld
6
Equation 1 2 x 3 8
5
18
2x 8
5

or
11
x
5
11 6
i.e. x & y
5 5
To find r , Let 2 x 3 y 8 be the regression equation of X on Y .

2x 8 3 y x 4 y
3
2
W
bxy Coefficient of Y in the equation of X on Y
Let 4 x y 10 be the regression equation of Y on X
y 10 4 x
3
2
TU
byx coefficient of X in the equation of Y on X 4 .
r bxy byx

3
4
2
bxy & byx are negative
2.45
JN

Since r is not in the range of 1 r 1 the assumption is wrong.

Now let equation 1 be the equation of Y on X
8 2x
y
3 3
byx Coefficient of X in the equation of Y on X
2
byx
ll

3
from equation (2) be the equation of X on Y
1
bxy
A

4
2 1
r bxy byx 0.4081
3 4
2
To compute y from equation 4 byx
3
 y
But we know that byx r
x
2 y
0.4081
3 2
y 3.26

ld
b). Marginal probability mass function of X is
1 3 4
When X 0, P X
8 8 8

or
2 2 4
X 1, P X
8 8 8
Marginal probability mass function of Y is
1 2 3
When Y 1, P Y
8 8 8

x
4
E X x p x 0 1
8
3
3 2 5
Y 1, P Y
8 8 8
4 4
8 8
5
E Y y p y 1 1
y 8 8
3 5 2
8 8 8
W
TU

E X 2 x 2 p x 02 12
4 4 4
x 8 8 8
E Y 2 y p y 1 12 1
2 2 3 5 3 5
y 8 8 8 8
V X E X 2 E X
2
JN

2
4 4 1

8 8 4
V Y E Y 2 E Y
2

2
1 15
1
4 16
E XY xy p x, y
ll

x y

1 3 2 2
0 0 1 1 0
8 8 8 8
A

1 1 1
Cov X , Y E XY E X E Y 0
2 4 8
 1
Cov X , Y
r 8 0.26 .
V X V Y 1 15
4 16

ld
Problem 27 a) Calculate the correlation coefficient for the following heights (in inches) of
fathers X and their sons Y .
X 65 66 67 67 68 69 70 72
Y 67 68 65 68 72 72 69 71
b) If X and Y are independent exponential variates with parameters 1, find the pdf of

or
U X Y .
Solution:

X Y XY X2 Y2
65 67 4355 4225 4489
66
67
68
69
68
65
72
72
W
4488
4355
4896
4968
4359
4489
4624
4761
4624
4285
5184
5184
TU
70 69 4830 4900 4761
72 71 5112 5184 5041

X 544 Y 552 XY 37560 X 2

37028 Y 2
38132

x 544
X 68
JN

n 8
y 552
Y 69
n 8
X Y 68 69 4692
1 2 1
X x2 X (37028) 682 4628.5 4624 2.121
n 8
ll

1 1
Y y2 y2 38132 692 4766.5 4761 2.345
n 8
1 1
Cov X , Y XY X Y 37650 68 69
A

n 8
4695 4692 3
The correlation coefficient of X and Y is given by
 Cov X , Y 3
r X ,Y
XY 2.121 2.345
3
0.6032 .
4.973

ld
b). Given that X and Y are exponential variates with parameters 1
f X x e x , x 0, fY y e y , y 0
Also f XY x, y f X x f y y since X and Y are independent
e xe y

or
e ; x 0, y 0
x y

Consider the transformations u x y and v y

x u v, y v v
x x
x, y u v 1 1
J
u , v y y 0 1
u v
1

W
fUV u , v f XY x, y J e x e y e u v e v
e , u v 0, v 0
u2v
RI R II
In Region I when u 0 v u
TU

f u f u, v dv e
u
.e 2 v dv u
u u

e 2 v
eu
2 u
eu eu
0 e 2u
JN

2 2
In Region II when u 0

f u f (u, v)dv
0

eu
e u 2v dv
0
2
ll

eu
2 , u 0
f u u
e , u 0
A

2
Problem 28 The joint pdf of X and Y is given by f x, y e
x y
, x 0, y 0 . Find the pdf of
X Y
U .
2
b) If X and Y are independent random variables each following N 0, 2 , find the pdf of
Z 2 X 3Y . If X and Y are independent rectangular variates on 0,1 find the distribution of

ld
X
.
Y
Solution:
x y

or
a). Consider the transformation u &v y
2
x 2u v and y v
x x
x, y u v 2 1
J 2
u , v y y 0 1

e
u v
fUV u , v f XY x, y J
x y
2 2e
2e2u , 2u v 0, v 0
u
x y
2e
2 u v v
W
fUV u , v 2e 2u , u 0, 0 v
TU
2
u u
2 2
f u fUV u , v dv 2e 2 u dv
0 0
u
2e 2u v 2
0
JN

u 2 u
2 e , u 0
f u 2
0 , otherwise

b).(i) Consider the transformations w y ,

i.e. z 2 x 3 y and w y
1
i.e. x z 3w , y w
ll

2
x x
1 3
x, y z w 1
J 2 2 .
A

z , w y y 2
0 1
z w
Given that X and Y are independent random variables following N 0, 2

x2 y2
1
f XY x, y e 8 , x, y
8
The joint pdf of z , w is given by
f ZW z , w J f XY x, y

ld
1
z 3 w w2
2
4
1 1
. e 8
2 8
1 321 z 3w2 4 w2
e , z , w .
16

or
The pdf of z is the marginal pdf obtained by interchanging f ZW z , w w.r.to w over the range of
w.
1

321 z 2 6 wz 13 w2
fZ z e dw
16

16

16
z 2 w2
e e
1 32

z
1 32
2

9z 2
13
32

6 wz 3 z 3 z

13
13

e 1332 e 32 13 dw


3
w
13

2
2

13
2

dw
W

TU
2

1 8z13 13
t2

16
e

e 32
dt

13 2 13 16 r 32
r t dr tdt dr dt dr dt
32 16 13t 13
1
16 13 4
dr dt r 2 dr dt
JN

13 r 32 13 2
z 1 2

2 4
e 813 e r r 2 dr
16 13 2 0
z 1 2

1
e 813 e r r 2 dr
2 13 2 0
z2
z2
ll

2 2 13
2
1 1
e 813
e
2 13 2 2 13 2

i.e. Z N 0, 2 13
A

b).(ii) Given that X and Y are uniform Variants over 0,1

 1, 0 x 1 1, 0 y 1
fX x and fY y
0, otherwise 0, otherwise
Since X and Y are independent,
1, 0 x, y 1
f XY x, y f X x f y y
0, otherwise

ld
x
Consider the transformation u and v y
y
i.e. x uv and y v
x x

or
x, y u v
J
u , v y y
u v
v 0
v
u 1
fUV u , v f XY x, y J
v , 0 u , 0 v
The range for u and v are identified as follows.
0 x 1 and 0 y 1. 0 uv 1 and 0 v 1
uv 0, uv 1, v 0 and v 1
W
TU
uv 0 and v 0 u 0
Now f u fUV u , v dv
The range for v differs in two regions
1
f u fUV u , v dv
0
JN

1
1
v2 1
vdv , 0 u 1
0 2 0 2
1
u
f u fUV u , v dv
0
1
1 u
u
v2 1
vdv 2 , 1 u
ll

0 2 0 2u
1
2 , 0 u 1
A

f u
1 , u 1
2u 2
Problem 29 a) If X 1 , X 2 ,..... X n are Poisson variates with parameter 2 . Use the central limit
theorem to estimate P 120 S n 160 where sn X 1 X 2 ...... X n and n 75 .
b) A random sample of size 100 is taken from a population whose mean is 60 and variance is
400. Using central limit theorem, with what probability can we assent that the mean of the
sample will not differ from 60 by more than 4.

ld
Solution:
a). Given that E X i 2 and Var X i 2
[Since in Poisson distribution mean and variance are equal to ]
i.e. 2 and 2 2

or
By central limit theorem, S n N n , n 2
S n N 150,150
120 150 160 150
P 120 Sn 160 P z
150 150
P(2.45 z 0.85)

W
P (2.45 z 0) P(0 z 0.85)
P(0 z 2.45) P(0 z 0.85)
0.4927 0.2939 0.7866

b). Given that n 100 , 60 , 2 400

TU
Since the probability statement is with respect to mean, we use the Linderberg-levy form of
central limit Theorem.
2 2
X N , i.e. X follows normal distribution with mean ' ' and variance .
n n
400
i.e. X N 60,
100
JN

X N 60, 4
mean of the sample will not X will not differ from
P P
differ from 60 by more than 4 60 by more than 4

P X 4
P 4 X 4
ll

P 4 X 60 4
56 60 64 60
P 56 X 64 P z
2
A

2
P 2 Z 2
2 P 0 Z 2 2 0.4773 0.9446
Problem 30 a) If the variable X 1 , X 2 , X 3 , X 4 are independent uniform variates in the interval
450,550 , find P 1900 X 1 X 2 X 3 X 4 2100 using central limit theorem.
b) A distribution with unknown mean has a variance equal to 1.5. Use central limit
theorem to find how large a sample should be taken from the distribution in order that the
probability will be at least 0.95 that the sample mean will be within 0.5 of the population mean.

ld
Solution
a). Given that X follows a uniform distribution in the interval 450,550
b a 450 550
Mean 500
2 2

or
b a 550 450
2 2

Variance 833.33
12 12
By CLT, S n X 1 X 2 X 3 X 4 follows a normal distribution with N n , n 2
S n n
The standard normal variable is given by Z

when S n 1900 , Z

when S n 2100 , Z
1900 4 500
4 833.33
2100 2000 100

4 833.33 57.73
100
57.73
1.732

1.732

P 1900 Sn 2100 P 1.732 z 1.732

W n 2
TU
2 P 0 z 1.732
2 0.4582 0.9164 .

b). Given E X i and Var X i 1.5

Let X denote the sample mean.
JN

1.5
By C.L.T, X follows N ,
n


We have to find ' n ' such that P 0.5 X 0.5 0.95

i.e. P 0.5 X 0.5 .95

P X 0.5 .95
ll

P z 0.5 0.95
n
n
A

P z 0.5 0.95

n
P z 0.5 0.95
1.5

ie P Z 0.4082 n 0.95
Where ' Z ' is the standard normal variable.

The Last value of ' n ' is obtained from P Z 0.4082 n 0.95

2 P 0 z 0.4082 n 0.95

ld
0.4082 n 1.96
n 23.05
The size of the sample must be atleast 24.

or
W
TU
JN
ll
A
 ld
or

W
TU



The analysis of univariate data in case of economic, social and scientic
sufcient. So, in some situations, say production price, height of father



JN

ll

ld

or

TU

JN

ll

ld

or
(1) Karl Pearsons Coefcient of correlation

moment correlation coefcient.

Karl Pearsons Correlation Coefcient,

W


mula, called Pearsons coefcient of correlation and denoted by

TU

JN

ll

ld
6. It is some how difcult to calculate.

The Karl Pearsons coefcient of correlation is based on the following

or

W







TU





When correlation is measured with help of Karl Pearsons coefcient o

JN

ll

ld
1. Find the means of the two series i.e., nd

3. Square these deviations and get the totals , i.e.,nd
and get the total i.e., nd

or

W
Calculate Karl Pearsons correlation coefcient from the following data

TU

JN

ll

Karl Pearsons Coefcient of Correlation,


A

Calculate coefcient of correlation between the marks obtained by 10 stud

ld

or

TU








JN

Compute Karl Pearsons coefcient of correlation from the following d

ll

Computing Table for Correlation Coefcient,



ld

or

W
and the coefcient of correlation between X and Y

TU

JN

Karl Pearsons Correlation Coefcient




ll

ld

or

W
Find the coefcient of correlation between the values of X and Y given be

TU

JN

ll






Then the correlation coefcient between and correlation coefcient between

ld
This implies correlation coefcient is independent of the change of origin

or

Net Prot (000 Rs.)

30
W
Total sales turnover and net prot of seven medium sized companies a
late the Karl Pearsons correlation coefcient:

50

60

80

100

110

130
TU

JN

ll


1. Calculate total for each series i.e., nd

ld

or
Find the coefcient of correlation between X and Y variables for the data g

W
















TU






JN

ll

ld

or

and 9 , nd the coefcient of correlation.

TU

A student while computing the correlation coefcient between two variables

JN

and nd the Karl Pearsons
correlation coefcient.
Subtract wrong gures from each total and add correct gure to get

ll


Calculate Karl Pearsons correlation coefcient

ld

efcient of correla-

or

Calculate Karl Pearsons coefcient of correlation between the ages o

W

months training of personnel. The following are the gures regarding se



Find correlation coefcient between

Calculate Karl Pearsons coefcient of correlation between advertise

TU
prot (Y) for 9 months:

Calculate the product moment coefcient of correlation between

JN

What will be the coefcient of correlation between

coefcient between

out the correlation coefcient:
ll

Calculate Karl Pearsons coefcient of correlation from the following d

A

Calculate Karl Pearsons coefcient of correlation between marks se

ld

Calculate correlation coefcient between age and death-rate from the da

or

Calculate Karl Pearsons correlation coefcient between age and litera

Calculate correlation coefcient between age and success in examination

TU








JN

Find Karl Pearsons correlation coefcient between age and playing h

ll

 Calculate the percentage of illiterate population from the following data and nd

ld

Calculate coefcient of correlation between density of population and dea

or

W














TU
From the following data calculate the coefcient of correlation by Karl Pea



Calculate Karl Pearsons correlation coefcient from the following data

JN

oefcient of

ll

A

But on subsequent verication, it was found that one pairs

ld
The interpretation of correlation coefcient depends on its degree and signicance. The correlation
coefcient lies between
The value of correlation coefcient is signicant or not is judged with the h

or




coefcient when various groups are used.

W



put them in the rst three columns respectively.
TU

may be and put them in rst three rows respectively.




JN

for each cell frequency and write the gures in the left hand upper

9. To nd





ll

Calculate the coefcient of correlation between ages of husbands and a



ld

or






















W

TU



[Table Interpretation: We rst obtain mid values of classes as

. For example for rst cell

last column for the rst row; there

JN

ll

 , correlation is signicant.

ld
Calculate coefcient of correlation from the following data:

or

W
The following gures are obtained in connection with the income and saving s

TU

JN

Calculate Karl Pearsons correlation coefcient from the following table

ll

Determine Karl Pearsons correlation coefcient from the following bi



ld

arranged in serial order (ranks), we nd correlation between the ran

or

or Ranking Method and the correlation coefcient so obtained
is called Rank Correlation Coefcient and is denoted by

Spearmans rank correlation coefcient is also used when the measur

However, Spearmans rank correlation coefcient

coefcient.

coefcient of rank correlation

W
coefcient between the ranks , it can be interpreted in the same way as the

TU




calculating rank correlation coefcient breaks down we use the following




JN

ll

ld

or

Use the rank correlation coefcient to discuss which pair of judges have

Here we calculate three rank correlation coefcient:




TU

JN

ll

ld
Find out the coefcient of correlation between

or

TU

JN

Find out the coefcient of correlation between X and Y by the method of ra

ll

ld

or

W










TU

JN

ll

ld

or
The coefcient of rank correlation of the marks obtained by 10 students


coefcient of rank correlation.

TU

JN

From the following data calculate Spearmans Rank coefcient of corre

ll

ld

2. It is the only formula to be used for nding correlation coefcient if we

or

TU

JN

Use Spearmans formula to nd the rank correlation coefcient.

ll

Calculate Spearmans rank correlation coefcient between ranks ( of the prot earned (X)

A

 Calculate the coefcient of correlation from the following data by Spearma

ld
Calculate rank correlation coefcient from the following data:

or
Calculate Spearmans rank correlation coefcient between advertisem

W
From the following data , nd the coefcient of correlation by the method of

Calculate correlation coefcient from the following data by rank differen

TU

Calculate correlation coefcient from the following data by Spearmans R

JN

Calculate correlation coefcient by the method of rank differences fro

ll

 ld
or

W
TU

ession was rst used by a

JN

More specically, regression analysis helps us understand how the typic

ll

variables are held xed.

A

ld

or

2. Correlation analysis has limited applications as it is conned only to the study

4. In correlation analysis, the correlation coefcient is a measure of de

TU



tionship is not sufciently

JN

The regression analysis may be classied as:

ll


The lines of best t expressing mutual average relationship between two

. If we have no reason or justication to assume as dependent variable a

ld



The concept of lines of best t is based on the principle of least square
-axis. When the deviations from the points to the line of best t are measure

or

. When the deviations from the points to the line of best t are measured ver

W




hown in the gure 1.

TU

Let r be the correlation Coefcient between X and Y,

is called the regression coefcient of X on Y and is denoted by


is called the regression coefcient of Y on X and is denoted by

JN

ll

is regression coefcient of
A

ld

or

is regression coefcient of

TU

JN

Coefcient of correlation between the prices of the two places +0.8.


ll

ld

or

W



























TU

JN

ll

ld

or

TU

JN

ll

ld

or

For grouped data regression coefcients are given as

Regression coefcient of X on Y

Regression coefcient of Y on X

TU

JN

ll
A


ld




Regression coefcient of

or







Regression coefcient of

TU











JN

ll








2. If correlation coefcient r between x and y is zero (r=0), the two reg

ld

4. The geometric mean of the two regression coefcients

or
coefcient





5. The arithmetic mean of the regression coefcients is greater than correlation coefcient. i.e.,

W


6. The regression coefcient are independent of the change of orig


7. If one of the regression coefcient is greater than unity, the other mu


8. Regression coefcient is the algebraic measure of the slope of the re
TU

JN

If the two regression coefcients are -0.9 and -0.5. nd out the value of correlation coefcient.


ll

ld

is negative (i.e. the sign of the two regression coefcients is not same),

or

, (b) the coefcient of correlation

TU

(b) Correlation Coefcient:

JN

Regression coefcient of

ll

Regression coefcient of



ld

or

W

The correlation coefcient between the prices of item in Jabalpur and Nar

ely and the coefcient of

TU

JN

ll

(i) Obtain the two regression equations and nd out the most probable valu

A

Coefcient of correlation between

ld

or

pecialized eld and to

er factory in a similar eld



TU

) for ve households is as follows:

JN

ll

 Calculate the coefcient of correlation between the values of exports and

ld

or

nd




TU
coefcient of correlation between marks in two subjects.







the correlation coefcient between them.
JN

Also nd out the ratio of the standard deviations of the two variables.


,
both regression coefcient and correlation coefcient between


(iii) Regression coefcient of
coefcient between

ll

coefcient of 6 and correlation coefcient between

coefcient of

A

ld

or

are multiple regression coef-

W


TU

= partial regression coefcient which measures the change in
= Partial regression coefcient which measures
JN

ll

 are partial regression coefcients,

After nding the values of

ld



is the square root of the regression coefcients
correlation coefcient is the square root of the product of two partial regression coefcients

or

TU

JN

ll

 quantities. After nding the unknown quantities, they are substituted in the reg

ld

or

W



















TU

JN

ll

ld

or

The values of regression coefcients:

TU

JN

ll
A




From the following data, nd regression equation:

ld

or

TU
JN
ll
A
 ld
or

W
TU

JN

A universe or population may be dened as an aggregate of items posse





ll

Innite


A population is said to be nite if the number of in-

dividuals is xed, i.e., nite. A population is said to be innite if it is composed of innitely large

ld
Students of your university constitute a nite population, leaves of a tree constitute an innite popu-
lation (here innite means very large).

A population is said to be hypothetical or articial if it is imaginary or it is constr

or

TU




The theory of sampling is concerned rst, with estimating the parameters of

JN

ll

deviation, correlation coefcient, etc. are called parameters or constan

ld

or

TU

JN

regularity and plays a signicant role in the sampling theory. This principle s

ll

ld

or

TU

level of signicance
When we select a particular level for signicance, say 5% and the proba
hypothesis is 95%. As we reduce the level of signicance, we reduce the
JN

of signicance is considered 5% or 1%.

ll

data are known as tests of signicance
 Testing of hypothesis and test of signicance are synonymous in some s

population parameter is signicant. If this hypothesis is accepted, we say th
signicant. The difference, if any, is due to sampling uctuations. The fo
steps) for test of signicances.

ld
1. Determination of the Problem: The rst and important function for a test of signicance is



2. Setting up of a Null Hypothesis: To know the signicance of difference

or

tween parameter and statistic; and if any it is due to chance or due to uctuatio

W
3. Selection of Level of Signicance: A predetermined hypothesis is teste
signicance. In general, we use 10% (or 0.01) and 5% (or 0.05) level of signicance.



, Standard Error for correlation coefcients,
TU

5. Computation of the Ratio Signicance: To nd a ratio signicance (often c

Ratio Signicance for mean

6. Interpretation: The last step of test of signicance is interpretation (or

JN

value of ratio signicance is more than critical value at 5% level of signica

96, we say that difference is signicant. This means the difference is no
to uctuations of sampling but there are other causes.

So far test of signicance

ll

methods of tests of signicance. There is no denite line between large sam



ld

or

TU

JN

ll

An investigator reports 1,700 sons and 1,500 daughters. Do these gures conrm to the hy-

ld

or

the 95% condence limits for the population.
TU

95% condence limits for the proportion:
JN

ll

ld
Condence limits for proportion :

or

W




TU

JN

ll

of the difference then the difference is signicant otherwise not signic


districts are signicantly different with respect to the prevalence of smok

ld

or

signicantly different with respect to the prevalence of smoking among m

TU


sample selected in the rst case was of 600 and in the second case it was 5




JN

ll


ence is signicant.

ld

or







is signicant otherwise not signicant.

W


difference signicant.


TU

JN

is signicant.

ll

this difference likely to be hidden at 5% level of signicance in samples of 1


A

ld

or

If level of signicance is taken as 5%, then 2 96, hence difference is signicant and it is not

TU

JN

ll

Limits for population mean at 0.27% level of signicance are : 172



ld
this range. Hence, then condence interval is

or

From a normally distribution innite number of iron bars with mean and standa
TU

JN

ll

n 10 to take sufcient


A

The difference at 5% level of signicance =

ld

or

TU

esis, The difference of the means of two samples is not signicant or th


onsidered signicant
otherwise not signicant.

JN

ll

is S.D. of rst sample and

A

ld

The mean produce of wheat of a sample of 100 elds comes to 200 lb. pe
deviation of 10 lb. Another sample of 150 elds gives the mean at 220 lb. w
iverse, nd out

or
if there is a signicant difference between the yield of the two samples.

standard error, the two mean yields are signicantly different.

TU

of Rs. 12, is the difference between the two means statistically signicant?

JN

ignicant.
ll

the standard deviation of the mean population of villages in the district is 10, n
mean of the rst sample is signicantly different from the combined mean



Standard Error of difference between the mean of rst sample and the

ld

or
of the standard error; hence it is signicant.

TU





ignicant i.e., the two samples

JN

Two samples of 100 and 80 students are taken with a view to nd out their ave
expenditure. It is found that median monthly expenditure for the rst gr
the second group is Rs. 100. The standard deviation for the rst group
ll

signicant.


A

ld

error; hence the difference is statistically signicant.

or

TU

be 13.9 and 17.6 respectively. If population standard deviation is 15,
JN

signicant difference between two standard deviation.

ll

of the standard error, hence it is not signicant.

A

From the following data, test the signicance of the difference of the stand


ld


signicant.

or


deviation of the rst sample differs signicantly from the combined standa

TU

The difference between standard deviations of rst sample and combin

of the standard error hence signicant.
JN

dence interval



ll

 signicant difference between town A and town B, so far as the proportio


bly except to nd in the whole

ld

percentage of deaf persons be attributed solely to uctuations of sampling

there any signicant

or
A random sample of 200 measurements from an innite population gave me
and standard deviation of 9. Determine the 95% condence interval for th

W


between the averages of the two samples statistically signicant?



The mean produce of wheat of a sample of 100 elds is 200 kilogram per
TU
standard deviation of 10 kilogram. Another sample of 150 elds gives the
the mean eld at 11
kilogram for the universe nd at 1% level if the two resuls are consistent.

JN

sample variance to nd the standard error of mean when population stand




ignicance can be used.



ll

ld
One of the most important test of signicance in case of small samples is

or

W
To test the signicance between the difference of sample mean and popu

1. Null Hypothesis: The population mean is denite, say mean = 50 cm

TU

JN

ll

: For relevant degrees of freedom (d.f.) at given level of signic

. For example, for 8 d.f. at 5% level of signicance the table value (fro

A

, then difference is signicant and over null hypothesis is false.


ld

or

777 is less than value with 5 degrees of freedom at 5% level of signicanc
TU

JN

For 9 d.f. at 5% level of signicance the value of

corresponding to 9 degrees of freedom at 5% level of signicance is 2

ll

ld
The condence limits for population mean are as follows:
95% condence interval for

99% condence interval for

or

mean? Obtain (a) 95%, and (b) 99% condence limits of the mean of the pop

TU

For 19 d.f. at 5% level of signicance,

On 95% Condence Interval: For 19 d.f.,

JN

ll

To test the signicance between the difference of two sample means taken
is dened as follows:
A

ld

or

W
s is signicant?
TU








Now, we shall nd out number of d.f. = (N-1) =
98 d.f. at 5% and 1% level of signicance are 1.984 and 2.626 respe
JN

of t is several times as large as these limits, hence there is a signicant diff

ll

Examine the signicance of the difference between the mean yields due to th
A

ld

or

Table value of t for 13 d.f. at 5% level of signicant = 2.16. Calculated
. Hence, the difference of average yields is not signicant

TU



group chosen at random. Test at 5% level of signicance whether there is signicance in the

JN

ll

Table value of t for 38 d.f. at 5% level of signicance = 2.02. Compute

. The difference the average income of two groups is signicant





-statistic is dened as follows with

ld



Null Hypothesis : The population correlation coefcient is zero.

or
It was found that the correlation coefcient between two variables calcu


Null Hypothesis : The population correlation coefcient is zero Given :

W






at 5% level of signicance = 2.069. Calculated value of

correlation coefcient is zero.
TU

ulation signicant




JN

12 at 5% level of signicance



ll

ld

or

TU

JN

months special coaching a second test of equal difculty was hold at the
marks give evidence that the students have been beneted by special c

At 5% level of signicance for 10 and 11 degrees of freedom, the table

ll

For 8 degrees of freedom the table value of t at 5% level of signicance





test to nd whether the two sets of data were drawn from populations with the s

ld
s. The coefcient

values signicant?

or

obtain according to a specic hypothesis. For example, if, according to M

might want to know about the goodness to t between the observed and

W


always testing what null hypothesis, which states that there is no signican

TU

Another way of dening the
JN

ll

lies in the rst quadrant.


ld

or

TU

level of signicance , then if

JN

ll

ld

or

W

Now we consider a test to determine if a population has a specied theoretica






TU

test whether the differences are likely to occur due to uctuation of samplin
JN

For explaining the goodness of t test, let us consider a population which m
classied into

ll

Now, a goodness of t test between observed and expected frequen



value will be small , which indicates that the t is good. On the other hand
will be large and t is not good. thus , acceptance of

ld
means the t is good and its rejection means that the t is not good.
level of signicance depends on the critical value of

or

1. In the goodness of t test observed frequencies in the k class are g
t is good.

TU

See that the expected frequencies in rst , second, fth and sixth class

JN

ll

ld

or

TU

JN

ll

ld

or

TU

JN

is much less than the table at 5% level of signicance , so

ll

ld

or

TU

JN

ll

 To nd the probability of this result we have, the degree of freedom=4-1

ld

or
Suppose N observations in a sample are to be classied according to two attr

. then the sample observations may be classied as shown:

W


































TU






JN

ll

classied. Let

A

ld

or


of signicance is taken as under:






level of signicance The value of

TU

The following table shows the classication of 4000 workers in a factory, a

JN

test to nd if there is any evidence to support that there was any associatio

of sesnicance.
ll

are independent. Assuming to be the hypothesis true, we nd the expected


ld

or

at 1 degree of freedom for 5% level of signicance is 3.841. Since ca

TU

JN

ll

 at 5% level of signication and conclude that

ld

or

TU


other got rst class. Are these gures commensurate with the general

JN
ll
A


1BasicsofQueuing
Queuingprocess: Queuingprocessisaclassofrandomprocessesdescribingphenomenaofqueue
formation.

CustomersArrivingServedCustomersLeaving

ld
SERVICE
FACILITY

or
Discouraged

Customersleaving
ATypicalQueuingProcess

W
Queueing theory: is the mathematical study of waiting lines (or queues). The theory enables
mathematical analysis of several related processes, including arriving at the (back of the) queue,
waiting in the queue (essentially a storage process), and being served by the server(s) at the front of
the queue.

1.1Characteristicsofaqueuingprocess
The following are the six basic characteristics of a queuing process:
TU

1. Arrival pattern of customers: In queuing the arrival process is usually stochastic. As a result it
is necessary to determine the probability distribution of the interarrival times (times between
successive customer arrivals) as well. Also customers can arrive in individually or
simultaneously (batch or bulk arrivals).
2. Service pattern of customers: As in arrivals, a probability distribution is needed for describing
the sequence of customer service time. Service may also be single or batch. The service
JN

process may depend on the number of customers waiting in queue for service. In this it is
called state dependent service.
3. Queue discipline: Queue discipline refers to the manner in which customers are selected for
service when a queue has formed. The default is FCFS i.e. is first come first served. Some
others are LCFS (last come first served), RSS (random service selection) i.e. selection for
service in random order independent of the time of arrival and there are other priority systems
where customers are given priorities upon entering the system, ones with higher priority are
selected first.
ll

4. System capacity: A queuing system can be finite or infinite. In certain queuing process there
is a limitation on the length of the queue i.e. customers are not allowed to enter if the queue
has reached a certain length. These are called finite queuing systems. If there is no restriction
on the length of the queue then it is called an infinite queuing system.
A

5. Number of service channels: A queuing system can be single or a multiserver system. In a

multiserver queuing system there are several parallel servers running to serve a single line.

A Multiserver Queuing system

 6. Number of service stages: A queuing system may have only a single stage of service. But as
an example of a multistage queuing system consider the physical examination procedure,
where each patient proceeds through various stages of medical examination, like throat check
up, eye test, blood test etc.

ld
Kendallsnotation

Aqueuingprocesscanbedescribedusinganotationwhichusesseriesofsymbolsandslashessuch
asA/B/X/Y/Z,where

or
A:indicatestheinterarrivaltimedistribution
B:indicatestheservicetimedistribution
X:thenumberofparallelservers
Y:therestrictiononsystemcapacity
Z:thequeuediscipline
StandardsymbolsforthecharacteristicsAandB

M
D
Exponential
Deterministic
W
TU
EK Erlangtypek
HK Mixtureofkexponentials
PH Phasetype
G General

1.2PoissonProcessandExponentialDistribution
JN

Themostcommonstochasticmodelsassumethatthearrivalrateandserviceratefollowapoisson
distribution.

ThePoissonprocessisacountingprocess{N(t),t0},whereN(t)denotesthetotalnumberofarrivals
uptotimetwithN(0)=0andwhichsatisfiesthefollowingthreeassumptions:

i. Theprobabilitythatanarrivaloccursbetweentimetandt+tisequaltot+o(t),
whereisaconstantindependentofN(t)and
ll

=0
ii. Pr{morethanonearrivalbetweentandt+t}=o(t)
A

iii. Thenumberofarrivalsinnonoverlappingintervalsisindependent,i.e.the
processhasindependentincrements.

Stationaryincrements:OneofthemostimportantpropertiesofPoissonprocessisthatthe
numberofoccurrencesinintervalsofequalwidthareidenticallydistributed.Inparticular,
fort>s,thedifferenceN(t)N(s)isidenticallydistributedasN(t+h)N(s+h).


Now,ifthearrivalprocessisPoissonthenitcanbeeasilyshownthattheassociatedrandom
variabledefinedasthetimebetweensuccessivearrivals(interarrivaltime)followsthe
exponentialdistribution.Likewiseiftheinterarrivaltimesareindependentandhavethe
sameexponentialdistribution,thenthearrivalratefollowsaPoissondistribution.

Theabovetheoryalsoholdsforservicerateandservicetimes.

ld
Markovianpropertyoftheexponentialdistribution:AstochasticprocesshastheMarkov
propertyiftheconditionalprobabilitydistributionoffuturestatesoftheprocessdepends
onlyuponthepresentstate;thatis,giventhepresent,thefuturedoesnotdependonthe

or
past. For service times this property states that the probability that a customer currently in service
has t units of remaining service is independent of how long it has already been in service. Thus we
have,

Pr{Tt1|Tt0} = Pr{0Tt1-t0}

distribution does not change when shifted in time or space i.e.

W
Stationary process (or strict(ly) stationary process) is a stochastic process whose joint probability

If Xt is a stochastic process then Xt is said to be stationary if, for all k, for all , and for all,
t1,t2,...tk
TU

1.3Markovprocess
Markov process, named after the Russian mathematician Andrey Markov, is a time-varying random
phenomenon for which a specific property the Markov property holds.

MarkovChain:Ifweassumethatthestatespace,I,isdiscrete,thentheMarkovprocessisknownas
JN

aMarkovChain

DTMC(DiscreteTimeMarkovChain):Iftheparametricspace,T,isalsodiscrete,thentheMarkov
chainisknownasadiscretetimeMarkovchain.InthiscaseweletT={0,1,2,...}.ForaDTMCthe
Markovpropertcanbestatedas

P(Xn=in|X0=i0,X1=i1,...,Xn1=in1)=P(Xn=in|Xn1=in1),i0,i1,...,inI
ll

CTMC(ContinuousTimeMarkovChain):Iftheparametricspace,T,iscontinuous,thentheMarkov
chainiscalledacontinuoustimeMarkovchain.InthiscaseweletT=[0,).ForaCTMCtheMarkov
propertycanbestatedas
A

P(X(t)=x|X(tn)=xn,X(tn1)=xn1,...,X(t0)=x0)=P(X(t)=x|X(tn)=xn)

1.4Birthdeathprocess
The birthdeath process is a special case of CTMC where the statesrepresent the current
sizeofapopulationandwherethetransitionsarelimitedtobirthsanddeaths.Birthdeath
processes have many applications in demography, queueing theory, performance
engineering,orinbiology.

When a birth occurs, the process goes from state n to n+1. When a death occurs, the

processgoesfromstatentostaten1.Theprocessisspecifiedbybirthrates

ld
anddeathrates .

or

Examples

A pure birth process is a birth-death process where i = 0 for all

A pure death process is a birth-death process where i = 0 for all

W .

.
TU
A (homogeneous) Poisson process is a pure birth process where i = for all

M/M/1 model and M/M/c model, both used in queueing theory, are birth-death processes
used to describe customers in an infinite queue.
JN

2MarkovianQueuingModels
2.1M/M/1/
TheM/M/1queuingsystemisdescribedasaqueuingmodelwhere:

arrivals are a Poisson process i.e. interarrival time is exponentially distributed;

ll

service time is exponentially distributed;

there is one server;
the length of queue in which arriving users wait before being served is infinite;
A

the population of users (i.e. the pool of users) available to join the system is infinite

321


AsimpleM/M/1queuewitharrivalrateandservicerate

2.1.1SteadyStateDistribution
Letpnrepresentstheprobabilitymassfunctionofadiscreterandomvariabledenotingthenumber
ofcustomersinthesysteminlongrun
pn=(1)n,<1

ld
where,
=/representsthetrafficintensityofthesystem.Forastablesystemtheintensitymustbeless
than1.
ItcanbeseenabovethatthesteadystateprobabilitiesforanM/M/1queuefollowsthegeometric

or
distributionwithparameter(1)

MeasuresofEffectiveness
Measure Expression
Averagenumberof
customersinthesystem(Ls)
Averagenumberof
customersintheQueue(Lq)
Expectedwaitingtimein
system(W)
/(1)

2/(1)

1/()
W
Expectedwaitingtimein
TU
/()
queue(Wq)
Utilization

2.1.2Transientsolution
Thetransientprobabilitiespn(t)=Pr{X(t)=n}foranM/M/1queuearegivenby
JN

ll

foralln0,where
A

istheinfiniteseriesforthemodifiedBesselfunctionofthefirstkind.

2.2M/M/1/N
ThissystemisatypeofM/M/1/queuewithatmostN(+veinteger)customersallowedinthe
system.
2.2.1SteadyStateDistribution
Thestateprobabilitiesinequilibriumaregivenby:

ld
MeasuresofEffectiveness

Measure Expression
Averagenumberofcustomers
inthesystem(Ls)

or
Averagenumberofcustomers Ls(/)
inqueue(Lq)
Expectedwaitingtimein Ls/
system(W)
Expectedwaitingtimein
queue(Wq)
Utilization
BlockingProbability(PB)
Lq/

W
TU

Throughput (1 PB )

2.3M/M/c/
Thissystemisamultiservermodelinwhichtherearecserversandeachserverhasanindependent
andidenticallydistributedexponentialservicetimedistribution,withthearrivalprocessagain
JN

assumedtobePoisson.

2.3.1SteadyStateDistribution
Forthismodelthesteadystateprobabilitiesaregivenby:
ll

where,
A

,<1

=/c,r==/

MeasuresofEffectiveness
Measure Expression
Averagenumberofcustomers
inthesystem(Ls)
Averagenumberofcustomers
intheQueue(Lq)
Expectedwaitingtimein
system(W)

ld
Expectedwaitingtimein
queue(Wq)

or
2.4M/M/c/K
2.4.1Steadystatedistribution
InthismodelthereisalimitKplacedonthenumberallowedinthesystematanytime.Thesteady
statesystemsizeprobabilitiesaregivenby:

W
Forthismodelthesteadystateprobabilitiesaregivenby:
TU

where,
JN

where, ,

MeasuresofEffectiveness
ll

Measure Expression
Averagenumberof ,1
customersinthe
A

Queue(Lq)
Averagenumberof Lq+r(1PK)
customersinthe
system(Ls)
Expectedwaitingtimein
system(W)
Expectedwaitingtimein
queue(Wq)
Utilization
BlockingProbability(PB)

Throughput

2.5M/M/c/c

ld
2.5.1SteadyStateDistribution
Thespecialcaseofthetruncatedqueuem/M/c/KforwhichK=c,i.e.wherenolineisallowedto
form.Forthismodelthesteadystateprobabilitiesaregivenby:

or

Measure
BlockingProbability(PB)
W
IncaseofanM/M/cmodelwedefinethefollowingperformancemeasures:

Expression
TU
Throughput (1 PB )
where,r=(/)

2.6BulkInput
2.6.1SteadyStatedistribution
JN

Thisisaqueuingmodelwhereininadditiontotheassumptionthatthearrivalprocessformsa
Poissonprocess,weassumethattheactualnumberofcustomersinanyarrivingmoduleisarandom
variableX,whichmaytakeonanypossibleintegralvaluelessthanwithprobabilityCx.Inthis
modelthebatchsizecanbeaconstantorarandomvariablewithsomedistribution.Belowwehave
discussedtwocases,sizeconstantorgeometricallydistributed.

Whenthebatchsizesaredistributedgeometrically
ll

where,
A

Measuresofeffectivenesswhenbatchsizeisconstant(K)

Measure Expression
Averagenumberof
customersintheQueue(Lq)
Averagenumberof
customersinthesystem(Ls)

Measuresofeffectivenesswhenbatchsizeisdistributedgeometricallywithparameter(1)

Measure Expression
Averagenumberof

ld
customersintheQueue(Lq)
Averagenumberof
customersinthesystem(Ls)

or
where, ,

2.7BulkService
ForthismodelitisassumedthatthearrivalsoccuratasinglechannelfacilityasanordinaryPoisson

W
process,theyareservedFCFSandthatthesecustomersareservedKatatime.Iflessthankarein
service,newarrivalsimmediatelyenterserviceuptothelimitK,andfinishwiththeothers
regardlessofthetimeintoserviceaftertheservicebegins.Alsotheamountoftimerequiredforthe
serviceofabatchisexponentiallydistributedregardlessofthefactthatthebatchisoffullsizeK.
TU
JN
ll
A
Arrivalrate():Averagerateatwhichcustomersarrivetothesystem.Hasunitsof"customers/
timeunit".

BlockingProbability:Blockingprobabilitygivestheprobabilityoftheeventthatanarrivalfindsall
serversbusyandleaveswithoutservice.

ld
FIFO:Firstin,firstout(FIFO)queuingisthemostbasicqueueschedulingdiscipline.InFIFOqueuing,
allpacketsaretreatedequallybyplacingthemintoasinglequeue,andthenservicingtheminthe
sameorderthattheywereplacedintothequeue.FIFOqueuingisalsoreferredtoasFirstcome,first
served(FCFS)queuing.

or
LIFO:TheLIFOQueuemodelsupportsthelastin,firstout(LIFO)queuingdiscipline.Theentitythat
departsfromthequeueatagiventimeisthemostrecentarrival.

M/M/1model:Aqueuingmodelwithoneserverandarrivalandservicetimeexponentially
distributed.

distributed.
W
M/M/cmodel:Aqueuingmodelwithcserversandarrivalandservicetimeexponentially

M/M/1/Nmodel:Aqueuingmodelwith1server,systemcapacityNandarrivalandservicetime
exponentiallydistributed.
TU
M/M/c/Kmodel:Aqueuingmodelwithcservers,systemcapacityKandarrivalandservicetime
exponentiallydistributed

Orbit:Queuingsystemswithretrialoftheattemptsarecharacterizedbythefactthatanarrival
customerwhofindstheserveroccupiedisobligedtojoinagroupofblockedcustomers,calledorbit,
andreapplyafterrandomintervalsoftimetoobtaintheservice.

JN

Queuingdelay:isthetimeajobwaitsinqueueuntilitcanbeexecuted.

Queuingmodel:isusedtoapproximatearealqueuingsituationorsystem,sothequeuingbehaviour
canbeanalysedmathematically.Queuingmodelshelpmeasureanumberofuseful steady
state performancemeasuresincludingtheaveragenumberinthequeue,orthesystem,average
timespentinthequeue,orthesystem,distributionofwaitingtimes,theprobabilitythequeueis
ll

full,orempty,andtheprobabilityoffindingthesysteminaparticularstate.

Retrialqueues:Manyqueuingsystemshavethefeaturethat,customerswhofindallserversbusy
uponarrivalareobligedtoleavetheserviceareaandcomebacktothesystemafterarandom
amountoftime

SamplePath:Asamplepathforthestochasticprocess{Xt:tT}isthefunctiononTtotherangeof
theprocesswhichassignstoeachtthevalueXt(w),wherewisapreviouslygivenfixedpointinthe
domainoftheprocess.
ServiceRate():Theaveragerateatwhichanindividualservercanservecustomers.Hasunitsof
"customers/timeunit"andisthereciprocaloftheaveragetimeittakestoserveonecustomer.

ServiceDiscipline:Queuedisciplineorservicedisciplinereferstothemannerinwhichcustomersare
selectedforservicewhenaqueuehasformedlikeFIFO/FCFS,LIFOetc.

Servers:Thenumberofserversavailabletoservecustomersenteringaqueuingsystem.Thenumber
ofserversmustbeawholenumberthatisgreaterthanorequaltoone.

ld
SteadyState:Thestateofthesystemafterithasbeeninoperationforalongtime.

SystemCapacity:Thetotalnumberofcustomersthatcanbeinthesystem,eitherwaitingorbeing
served.Mustbeawholenumberthatisgreaterthanorequaltoone.

or
TrafficIntensity:Trafficintensityisameasureoftheaverageoccupancyofaserverorresource
duringaspecifiedperiodoftime.

Throughput:isthenumberofcustomersservedperunittime

W
Utilization:Theproportionoftimeaserverisbusyistheserverutilization.

WaitingTimes:Customerwaitingtimecanbeoftwotypes,thetimethecustomerspendsinthe
queueandthetotaltimeacustomerspendsinthesystem(waitingtimeinqueue+service).

TU

JN
ll
A
 ld
Problem 1 For M / M /1 : / FIFO model, write down the Littles Formula.

or
Solution:
i) LS WS
ii) Lq Wq
1
iii) WS Wq

iv) LS Lq


W
Problem 2 For M / M / C : N / FIFO model, write down the Formula for
a) Average number of customers in the queue.
b) Average waiting time in the system.
TU

Solution:

a) Lq 0 1 n c n c 1 n c when
C !1 c
1
b) WS 1 LS

JN

i 1

1 C c n n
n0
1
LS Lq

Problem 3 What is the probability that a customer has to wait more than 15 minutes to
get his service completed in a M / M /1 queuing system, if 6 per hour 10 per
ll

hour?

Solution:
Probability that the waiting time in the system exceeds t is
A

e
W
dW e t
t
 15 10 6. 14
P WS e e 1 0.3679
60
Problem 4 What is the probability that an arrival to an infinite capacity 3 server Poisson
1
queueing system with 2 and P0 entries the service without waiting.
9

ld
Solution:
P[Without waiting] P N 3 P0 P1 P2
n
1
Pn P0 when n c 3 .
n!

or
1 2 1 1 5
P N 3 22 .
9 9 2 9 9
Problem 5 In a given M / M /1 : / FCFS queue, 0.6 , What is the probability
that the queue contain 5 or more students?

Solution:
P X 5 5 0, 6 0.0467
Problem 6
when 2 and 5
5

W
What is the effective arrival rate for M / M /1 : 4 / FCFS queuing model
TU
Solution:
1 1
1 / 1 2 / 5
5 1 0
1 / 1 2 / 5
K 1 5

3/ 5 0.6
5 1 5
5 1
JN

1 2 / 5 1 0.01024
0.6
5 1 5 1 0.607
0.989
5 0.393 1.965 2
Problem 7 a) In M / M /1 : K / FIFO write down the expression for P0.
b) In M / M /1 : K / FIFO , 3/ hr , 4 / hr , K 7 Calculate P0
ll

Solution:

1

a) P0 of
A

K 1

1

1
of
K 1


b) P0 1 K 1
of

1

3

ld
1 4 0.2778
8
3
1
4
Problem 8 In M / M / S : / FIFO , 10 / hr , 15 / hr , S 2 Calculate P0

or
Solution:
1
S 1 1 n 1 s s
P0
n 0 n ! S ! s

1 10 1 10 2 30
1
1! 15 2! 15 30 10
1
2
W
1

Problem 9 In a two server system with infinite capacity and 20 / hr and 11/ hr ,
TU
find P0.

Solution:
1
S 1 1 n 1 s s
P0
n 0 n ! S ! s
1
20 1 20 2 11 2
JN

1
11 2 11 22 20
1

21
Problem 10 Write the steady states equations in M / M /1 : K / FIFO where
n for n 0,1, 2...K 1
0 for n k
ll

and n n 1, 2,3

Solution:
A

Pn1 Pn Pn 1 1 n K 1
P1 P0 n 0
 Problem 11 In M / M /1 : K / FIFO , 3 / hr , 5 / hr , K 7 , what is the
probability that the server will be idle.

Solution:
The probabilities that the server will be idle

ld
1 3 2
1
5 5 0.4
P0 K 1
0.407
3 1 0.0168 0.983
8

1 1
5

or
Problem 12 In M / M / S : K / FIFO , write the expression for P0.

Solution:
1
s 1 1 1 k

ns

P0 / /

W
n s

ns s
n 0 n ! s!
Problem 13 What is effective arrival rate in M / M / S : K / FIFO

Solution:
Effective arrival rate
TU
s 1
1 s s n n
n 0
Problem 14 In M / M / S : K / FIFO with
3.76 / hr and 4, s 2, k 7, compute P5 where P0 0.3617 .

Solution:
JN

1
n s
/ P0
n
P5
s !s
5
1 3.76
0.3617
2!25 2 4
0.0166
ll

Problem15 A repairman is to be hired to repair machines which break down at an

average rate of 3 per hour. The breakdown follows Poisson distribution. Non- productive
A

time of machine is considered to cost Rs.16/hr. Two repairmen have been invited. One is
slow but cheap while the other is fast and expensive. The slow repairman charges Rs.8
per hour and he services machines at rate of 4 per hour. The fast repairman demands
Rs.10 per hour and service at the average rate of 6 per hour. Which repair man should be
hired?
 Solution:
For the slow man: Model M / M /1 : / FIFO 3 / Hr , 4 / hr
1 1
Expected waiting time of a machine 1hr
43
Non productive cost =16 / hr

ld
Non productive cost = 3 16 =Rs.48.
Charges paid to the repairman 8 3 24
Total cost =48+24=72
For fast man: M
3 / hr , 6 / hr.

or
1 1
Expected waiting time hour
3
1
Non productive cost 3 16 16
3
Charge paid to repairman 10 1 10
Total cost =16+10=26
Fast repairman can be hired.
W
Problem 16 A two person barber shop has 5 chairs to accommodate waiting customers.
Customers who arrive when all 5 chairs are full leave without entering barber shop.
Customers arrive at the average rate of 4 per hour and spend an average of 12 minutes in
TU
the barbers chair. Compute P0 , P1 and average number of customers in the queue

Solution: This is model M / M / C : K / FIF 0 4 / hr , 5 / hr , C 2, N 2 5 7

i
1 1 1 7
1
P0 / n2
/
n n

n0 n ! C! n2 2
1
4 8 2 2 2
JN

2 5

1 1 ...
5 25 5 5 5
0.4287
1
Pn n c / P0 , n 0
C C!
1
P7 7 2 4 / 5 0.4287
7

2 2!
ll

0.0014.
PC 1 N C 1 N C N C
C

Lq P0 2
C !1
A

Where / C
0.4
0.4287. 0.8 1 0.4 5 5 0.6 0.4 5 0.15 Customers
2

2!1 0.4
2
 Problem 17 Derive the formula for
i) Average number Lq of customers in the queue.
ii) Average waiting time of a customer in the queue for
M / M /1 : / FIFO MODEL.

ld
Solution:

Lq n 1 Pn nPn Pn
n 1 n 1 n 1


nP n 0 nPn 1 P0
P P

or
n
n 1 n 0 n 0
Ls 1 P0
2
1 1

Wq W

0

W
ii) Average waiting time of a customer is the queue Wq

e w dw where / e w is the probability density

function of the waiting time distribution of the queue.

w.e w dw

TU
0

e w

w
e dw
W
0 0



1


JN

Problem 18 On average 96 patients over 24 hour day require the service of an

emergency classic. Also on average a patient requires 10 minutes of active attention.
Assume that the facility can handle any one emergency at a time. Suppose that it costs the
clinic Rs.100 per patient treated to obtain an average servicing time of 10 minutes and
that each minute of decrease in this average time would cost Rs.10 per patient treated.
How much would have to be budgeted by the clinic to decrease the average size of the
1 1
queue from 1 patient to patient?
3 2
ll

96
Solution: This is M / M /1 : / FIFO model 4 / hr .
24
6 / hr
A

4 1
Lq . 1
3 3
1 1
Now 1 customers to be reduced to , then we have to find 1 such that
3 2
 1
ie 2 4 1 32 0
2
1
8 1 4 0 1 8 Patient /hr
1 15'
Average time required to each patient hr
8 2

ld
15 5
Decrease in the time required to attend a patient 10 min.
2 2
Decrease in each minute cost Rs.10 per patient.
5
Cost 10 25 .

or
2
Budget required for each patient= 100+25=125 to decrease the size of the queue.

Problem 19 Obtain the expressions for steady state probability of a M/M/C queuing
system

Solution:
n n 1 n C
C n c
W
It can be treated as a birth death process with n for all n and n
is given by (1)
TU

In a birth death process

1
n
i 1
i 1
n

a
Pn P0 and P0 1 for M/M/C.
i 1 i n 1 i
i 1

Pn 0 1 .... n 1 P0
JN

1 2 n 1
n
... n if 1 n C
P0

1 2
n
of 1 n c
ni n P0
n c 1 ...... c
Pn .... c 1 if n c
ll

1 2 n c 1
n
.... c 1 c 1
n c 1

1 2
A

n
C n1 of n c
C !n
 C 1 n
n
P0 nc n
n 0 n ! n n c c c !

Problem 20 Arrivals at a telephone booth are considered to be Poisson with an average

time of 12 mins. The length of a phone call is assumed to be distributed exponentially

ld
with mean 4 min. Find the average number of persons waiting in the system. What is the
probability that a person arriving at the booth will have to wait in the queue? Also
estimate the traction of the day. Where phone will be in use

Solution: This is model M / M /1 : / FIFO

or
1
12, 12 per min

1 1
4, per min
4

LS

1 1
1
12 0.5.

4 12 W
P LS 0 1 P t s 0 1 P [No customer in the system]
1
TU
1 P0 1 1 / / .
3
2
P Phone will be idle = P0 1
3
1
P Phone will be in use =
3
Problem 21 There are three typists in an office. Each typist can type an average of 6
JN

letters per hour. If letters arrive for being typed at the rate of 15 letters per hour, what
fraction of time all the typists per hour will be busy? What is the average number of
letters waiting to be types?

Solution: This is M / M / C : / FIFO

C 3, 15 / hr , 6 / hr
P [all the three typists buy] P N 3
ll

3

P0

A

3! 1
3
 1
P0 r 1
1 1
n ! n
n 0

3

c !1
c

ld
1


2.5
2
1 2.53
1 2.5
2!
3! 1 15
6 3

or
0.0449.
P N 3 0.7016
c 1


Lq

c.c !1

2.5
25
2

3 6 1 2
3
P0

0.0449 3.5078
W
TU

Problem 22 A bank has two tellers working are savings account. The first teller handles
with drawals only the second teller handles deposits only. It has been found that the
service time distributions for both deposits and withdrawals are exponential with mean
service time of 3 minutes per customer. Depositors are found to arrive in a poisson
fashion through out the day with mean arrival rate of 16 per hour withdrawals also arrive
in a poisson fashion with mean arrival rate of 14 per hour. What would be the effect on
JN

the average waiting time for the customers if each teller could handle both withdrawals
and deposits?

Solution:
1. When the tellers handle separately withdrawals and deposits. This is Model
M / M /1 : / FIFO .
For depositors 16 / hr , 20 / hr
ll

1
Wq hr 12 min
5
For withdrawals 14 / hr , 20 / hr
A

Wq 7 min

ii) If both tellers do both service, one it will be a M / M /1 : / FIFO model.
C2 20 / hr 16 14 30 / hr
 1
i 1 1 n 1 c c
P0
n 0 n ! c ! c
1
30 1 30 2 40 1
1
20 2i 20 40 30 7

ld
c

2
30
1
1 1 1 1 20
Wq P0
c.c ! 2
20 2.21 30 2 7
1 1

or
c 40
0.0642 hr or 3.857

Problem 23 In a heavy machine shop, the overhead crane is 75% utilized. Time study
observations gave the average slinging time as 10.5 minutes with a SD of 8.8 minutes.

W
What is the average calling rate for the service of the crane and what is the average delay
in getting service? If the average service time is cut to 8.0 minutes, with a standard
deviation of 6.0 minutes, how much reduction will occur on average in the delay of
getting service?

Solution: This is model M / G /1 : / FIFO

Average delay is getting service
TU

1 2 2 p /
2 1
0.75, 5.71/ hr , 8.8.
0.75 2 8.8
2
60
1 5.71
2 1 0.75 60 5.71
JN

Average calling rate 0.75 5.71 4.28 per hr. If the service time is cut to
60 4.29
8 minutes, then 7.5 / hr , 0.571
8 7.5
utilization of the crane reduced to 57.1 percent and average delay in getting service

0.571 2 600
2
60
Wq 1 7.5
2 1 0571 60 7.5
ll

8.3 minutes.
a reduction of 18.5 minutes approximately 70%
A

Problem 24 A Super market has two girls ringing up sales at the customers. If the service
time for each customer is exponential with mean 4 minutes and if the people arrive in a
Poisson fashion at the rate of 10 per hour
a) What is the probability of having to wait for service?
b) What is the expected percentage if idle time for each girl?
 Solution:
This is model M / M / C : / FIFO
1 1 1
, ; C 2 .
6 4 C 3
1
c 1 1 4 n 1 4 2 2.1/ 4 1
P0

ld

n 0 n ! 6 2! 6 2.1/ 4 1/ 6 2
2 1 1
P1 P0 .
3 2 3

1 1 1

or
a) P C 2 Pc 1
c2 2 3 6
Expected number of idle girls
b) Probability of any girl being idle .
Total number of girls
1 1
2 1

2 P0 1 P1
C
2
2 6
W
3 4 0.67

Hence expected percentage of idle time each girl is 67%.

Problem 25 At a one man barber shop, the customers arrive following poisson process at
an average rate of 5 per hour and they are serviced according to exponential distribution
with an average service. Rate of 10 minutes assuming that only 5 seats are available for
TU
waiting customers find the average time a customer spends in the system.

Solution:
This is model M / M /1 : N / FIFO
5 / hr.
1
60 6 per hour N 5
JN

10
5
P
6
5
1
1 6 0.1666 0.2505.
P0
1 N 1
5
6
0.6651
1
6
ll

L
WS S where


n 1
5
6
A

n 1 6
5 6
LS
65 5
6
1 n 1 1
6

 6.0335 2.01
5 5 198
1 0.335 0.665
' 1 P0
6 1 0.2505 4.497
1 1.98

ld
Ws LS 0.441 hrs
'
4.497
Problem 26 At a railway station, only one train is handled at a time. The railway yard is
sufficient only for two trains to wait while the other is given signal to leave the station.
Trains arrive at the station at an average of 12 per hour. Assuming poisson arrivals and

or
exponential service distribution, find the steady state probabilities for the number of
trains in the system also find the average waiting time of a new train coming into the
yard. If the handling rate is reduced to half, what is the effect of the above results?

Solution:
This is Model M / M /1 : K / FIFO
Hence 6 / hr , 12 / hr , k 3

Pn P0 when P0

1

W
Steady state probability for the number of trains in the system P1 , P2 and P3 .

K 1

1
TU

6
1
12 0.5 0.5 0.5333
1 0.5
4 4
6 0.9375
1
12
1

JN

P1 P0 0.2667

2

P2 P0 = 0.1334

3

P3 P0 0.0667

1
ll

Average waiting time of a new train coming in the yard is Ws LS where

K 1

K 1
A

Ls K 1

1

 4 0.5
4
6
0.7333.
12 6 1 0.5 4
1 1 c 12 1.05333 5.6004
0.7333
Ws 0.1309 hrs

ld
5.6004
Case II

If the handling rate is reduced to half, then 6 / hr , 6 / hr and k 3

1 1

or
Hence Pn
k 1 4
1
P1 P2 P3
4
Average waiting time of a new train coring into the yard in
L
Ws 1s

Ls
K

2
1
3
2
1 1 P0 6 1 4.5
4
W
Since 1.5 train

L 1.5
TU
Ws s1 0.333 hrs (or 201 )
4.5

Problem 27 At a port there are 6 unloading berths and 4 unloading crews. When all the
berths are full, arriving ships are diverted to an overflow facility 20 kms down the river.
Tankers arrive according to a poisson process with a mean of 1 for every 2 hrs. If it takes
for an unloading crew on the average, 10 hrs to unload a tanker, the unloading time
follows an exponential distribution Determine.
JN

a) How many tankers one of the port on the average?

b) How long does a tanker spend at the port as the average?

Solution:
1

P1 P0 0.0361

2
1
ll

P2 P0 0.0901
2!
3
1
A

P3 P0 0.1502
3!
Ls 4.8214 4 0.00721 3 0.0361 2 0.0901 0.1502 4.3539 tankers.
b) Time spend by a tanker at the port
 1 c 1

WS

L
1 S
Where 1


C
n 0
c n n

1 S 4 P0 3P1 2 P2 P3
1
4 4 0.00721 3 0.0361 2 0.0901 0.1502
10

ld
0.3532
4.3539
WS 12.3250
0.3532
Problem 28 A petrol pump has 2 pumps. The service time follows the exponential

or
distribution with a mean of 4 minutes and ears arrive for service in a poisson process at
the rate of 10 cars per hour. Find the probability that a customer has to want for service.
What proportion of time the pumps remain idle?

Solution:
This is model M / M / C : / FIFO

10 / hr

P0
c
1 4

60
is 15 / hr C 2

P[ a customer has to wait ] P n c

W

TU

C !1
c
2
10

P n 2
15
P0 where
10
2!1
JN

30
1
c 1 1 n 1 c c
P0
n 0 n ! c ! c p
1
1 10 1 10 2 30
1
1! 15 2! 15 30 10
ll

1
2 1 1
1
3 3 2
2
2
A

1 1
Probability that a customer has to want in P n 2 . 0.1667
3
2 2 6
2.
3
 10 1
The traction of time when the pumps are busy =traffic intensity
c 30 3
Hence the pumps remain idle 1
1 2
1 67%
3 3

ld
Problem 29 A supermarket has two girls serving at the customers. The customers arrive
in a Poisson fashion at the rate of 12 per hour. The service time for each customer is
exponential with mean 6 minutes. Find
i) The probability that an arriving customer has to want for service.
ii) The average number of customers in the system, and

or
iii)The average time spent by a customer in the supermarket.

Solution: This is M / M / C : / FIFO . Here

1 6
X 12 / hr , 10 / hr , c 2
60

P n c

Ci 1
C

P0

c


W
i) Probability that an arrival customer has to wait is
TU
12 1 12 2 20 1 1
P0
10 2! 10 20 12 4

2
12 1

P n 2
10 4
0.45
12
JN

2!1
20
The probability that an arrival customer has to want=0.45

c 1


1
ii) Average number of customer is the system Ls P
2 0
c.c !
1 c
ll

3
12
1 10 1 12
A

1.875 2 customer
2.2! 12 2 4 10
1
20
iii) Average time spent by a customer in the supermarket =Average waiting time
of a customer in the system
 1 1.875
Ws Ls 0.1563 hours 9.375 minutes.
12

Problem 30 Four counters are being run on the frontier of a country to check the
passports of the tourists. The tourists choose a counter at random. If the arrival at the

ld

frontier is poisson at the rate and the service time is exponential with parameter ,
2
find the steady average queue at each counter.

or
Solution:
This is M / M / C : / FIFO
Average queue length


1 P
Lq
c.c !
1 c

P0

2 0

3 1 n 1 c c
W

n 0 n ! Ci c
1
1
TU
2 2 23 1 2
1 2 24
2 6 4! 2
3

23
1 25 3 4
Lq 2
0.1739
4.4! 23 23
1
JN

Problem 31 In a given M/M/1 queuing system the average arrivals is 4 customers per
minutes 0.7 what are
1) Mean number of customers Ls in the system
2) Mean number of customers Lq in the queue.
3) Probability that the service is idle.
ll

4) Mean waiting time Ws in the system.

Solution:
A

0.7 0.7
Ls 2.33.
1 1 0.7 0.3
Lq Ls 2.333 0.7 1.633.
P0 1 1.07 0.3
 Ls 2.333
Ws 0.583
4

Problem 32 A TV repairman finds the time spend on his job has an exponential
distribution with mean 30.If he repairs sets in the model in which they come in and of
the arrival of sets is approximately poisson with an average rate of 10 sets per 8 hours

ld
day, what is the repairmans expected idle time each day ? How many jobs ahead of the
average set just brought in?

Solution:
10 5 1

or
We are given sets per hour 60 2 sets per hour
8 4 30
5
Then
8
5 3
The probability of no unit in the queue is P0 1 1

Ls

2
5

5
2
W
4 1 jobs = 2 TV Sets (approx)
3

3
8 8
Hence the idle time for repair man in an 8-hours day 8 3 hrs
8
TU
4

Problem 33 If for a period of 2 hours in the day trains arrive at the yard every 20
minutes but the service time continuous to remain 36 minutes. Calculate the following for
the above period.
i) The Probability that the yard is empty
ii) The average number of trains on the assumption that the line capacity of the yard
JN

is limited to 4 trains only.

Solution:
1 1
This model is M / M /1 : K / FIFO Hence per min; per min ,
20 36
K 4.
i) Probability that the yard is empty

1
ll

P0 K 1
,

1

A

36
1
20 0.0447
5
36
1
20
 ii) Average number of trains
K 1

K 1
LS

of
K 1

1

ld
5
1 36
5
5
20 20
30 27
1 1 36
1
36 20
20

or
=3 trains.

Problem 34 Suppose there are 3 typists is a typing pool. Each typist can type an average
of 6 letters / hr. If letters arrive to be typed at the rate of 15 letters/hr
a) What is the probabilities that there is only one letter in the system

Solution:
This is M / M / S : / FIFO
s 1 1 n 1 s s
W
b) What is the average number of letters waiting to be typed?
c) What is the average time a letter spends in the system?

1
TU
P0
n 0 n ! s ! s
1
2 1 15 n 1 15 3 18

n 0 n ! 6 3! 6 18 5
1
1 18
1 2.5 3.125 15.625
JN

6 3
0.04494.

n
1
a) P1 P0 Since n s
n!
1
1 15
0.04494 0.11235
1! 6
ll

b) Average numbers of letters waiting to be typed

S 1


1
A

Lq P
2 0
ss !
1 s

 4
15
1 6
0.04494
3.3! 15 2
1
18
3.51094 Letters.

ld
1
C) Ws Ls

1
Lq

or
1 15
3.51094 0.40073 hr
15 6
24.0438 .

Problem 35 An automobile inspection station has 3 inspection stalls. Assume that cans

W
wait in such a way that when a stall becomes vacancy the car at the head of the time pulls
up to it. The Station can accommodate at most 4 cars waiting at one time. The arrival
pattern is poisson with a mean of I car every minute during the peak hours. The service
time is exponential with mean 6 min. Find the average number of customers is the queue
during peak hours and the average waiting time in the queue.

Solution:
TU

This is M / M / C : K / FIFO Model.

c 3, k 7.
60 / hr 10 / hr.
1
i 1 1 m 1 C k m c
P0
m 0 m ! C ! m v c
JN

1
2 1 1 7
6
m 3

6 6
m 3

m 0 m ! 3! m 3 3
1
1 1
1 6 62 63 1 2 22 23 2 4
2! 3!
7 18 36 31 1141 0.000.8764
7 1
ll

i) Average no of customer is the queue.

c
1
Lq P0 1 K C K C 1 K C
A

2
c ! 1
6
Where 2
c 3
 2 1
63.0.0008764 1 24 4 1 2 24
2
3! 1 2
3.09
ii) Average waiting time in the queue.

ld
1 c 1

Wq

L
' q
where '
c
m 0
c m Pm

2

1 10 3 3 m Pm
m 0

or
10 3 3P0 2 P1 P2
10 3 3 0.00087 2 0.005256 0.015768
29.76 30
3.09
Lq 0.103hrs.
30

1 1
1
n

Hence Pn P0 ; n c 3
n!

P1 6 0.0008764 0.005256
1!
1
W
P2 6 0.00087 0.015768
2
TU
2!
Problem 36 Patients arrive at a clinic according to Poisson distribution at a rate of 60
patients per hour. The waiting room does not accommodate more than 14 patients
Investigation time per patient in exponential with mean rate of 40 per hour.
a) Determine the effective arrival rate at the clinic.
b) What is the probability that an arriving patient will not wait?
c) What is the expected time (waiting) until the patient is discharged from the
JN

clinic?
Solution: This is M / M / I : K / FIFO model hence 60 patients / hr,
40 / hr. K 14 1 15
a) Effective arrival rate
1 1 P0

1

ll

Where P0 K 1
of

1

A

60
1
40 0.0007624
6
60
1
40
 1 40 1 0.0007624 39.9695 / hr
b) An arrived will rut want =there is no patient is the clinic.
P[ a patient will not wait ] P0 0.0007624
c) Effective waiting time
L
WS S1

ld

K 1

K 1
When LS

K 1

or
1

16
60
16
60 40

40 60 60
16

Ws
39.9695
40
2 16.00 14 Patient
14
0.3503 hr or 21.018 W
Problem 37 A group of users in a computer browsing centre has 2 terminals. The
TU
average computing job requires 20 min of terminal time and each user requires some
computation about once every half an hour .Assume that the arrival rate is poisson and
service rate is exponential and the group contains 6 users. Calculate
a) The average number of users waiting to use one of the terminals and in the
computing job.
b) The total time lost by all users per day when the centre is opened 12 hrs/day.
JN

Solution:
1 30 1 1 20 1
This is M / M / S : K / FIFO Hence ;
60 2 60 3
2 / hr , 3 / hr s 2, k 6
a) The average number of users waiting to use terminals and in the
Computing job is nothing but me average number of uses in the system.

S 1
LS Lq S S n Pn
ll

n 0
1
S 1 1 n 1 s k n s
P0
A

n 0 n ! S ! n s s
1
2 1 2 2 1 1 2 1 3 1 4
1 1 0.5004
3 2! 3 3 3 3 3
 s

Lq P0 1 y K S K S 1 k s Where y
2
S !1 s
1
3 . 1 1 4 2 1
2 4 4

2
0.5004 2
3 2 3 3 3

ld
2!
3
0.0796
s 1
LS Lq S S n Pn becomes

or
n 0

Lq S 2 P0 P1

Lq S 2 P0 P0

2
0.0796 2 2 0.5004 0.5004
0.7452 .

Wq
1
1
Lq where
3

W
b) Average waiting time of a user in the queue Wq

S 1

TU
1 S S n Pn
n 0
2
S 2 P0 P1 3 2 2 0.5004 0.5004
3
1.9968
1 0.0796
Wq L 0.0399 hr
JN

1 q
1.9968

Problem 38 The railway marshalling yard is sufficient only for trains (there are 11
lines one of which is earmarked for the shutting engine to reverse itself from the crest of
the hump to the rate of the train). Train arrive at the rate of 25 trains per day, inter-arrival
time and service time follow exponential distribution with an average of 30 minutes
Determine.
a) The probability that the yard is empty.
ll

b) The average queue length.

Solution:
A

This is model M / M / : K / FIFO

1
Hence 25 trains /day, 30 0.0333 train /min

0.0174
 K 10 11 1
a) Probability that the yard is empty.

1

P0 K 1
if

ld
1

0.0174
1
0.0333 1 0.5225 0.4779
1 0.5225
11 11
0.0174
1

or

0.0333
c) Average queue length.
K 1

K 1
LS
of
K 1

0.0174
0.0333 0.0174

1

11

0.0174

0.0333
0.0174
1
11
11
W
0.0333
TU
0.008721
1.09434 .
0.9992
1.0856
JN
ll
A
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld
A
ll
JN
TU
W
or
ld