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
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
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
definition.
Definition of Random Variable A random variable is a function from a sample space S into
the real numbers.
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
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
S = {s1,...,sn}

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

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,

All
PX(X = xi) = P({sj ∈ S : X(sj) = xi}). (1) 1
 JNTU

World
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
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.
(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(∪ mi=1{sj ∈ S : X(sj) = xi}) = P(S) = 1. (3) If A1,A2,... ∈ B and pairwise

disjoint then
PX(∪ ∞k=1Ak) = P(∪ ∞k=1{∪ xi∈ Ak{sj ∈ S : X(sj) = xi}})
= Pwhere the second inequality A note on notation: Random the realized values of the
 A l
Thus, the random Example 1.4.3 (Three three in space the The times is

l J
three range independently. tosses. X(s) for s variable A coin

N
variable Define tosses-II) X follows ∑k=1
 T
∞can variables the take will from Consider random the be will the denoted value

U
always fact variable again x.

W
P by be is the X a the denoted probability to experiment corresponding be
 o
the with number function. uppercase of X(Atossing of lowercase k), heads D

rld
letters a complete HHH 3 the random and

letters.
fair coin
obtained
enumeration of the value of X for each point in the sample
HHT HTH THH TTH THT HTT TTT
2221110
variable X is X = {0,1,2,3}. Assuming that all eight points
, by simply counting in the above display we see that the
in S have probability 18
induced
probability function on X is given by

2 P(∪ xi∈ Ak{sj ∈ S : X(sj) = xi} = ∑∞k=1

x0123
 All
PX(X = x) 18 ⎛38

JNT U
⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎨ 0 18 if −∞

<x< if 0 ≤ x < 1

W
0

⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎝1 12 78 if 1 ≤ x < 2 if 2 ≤ x < 3
if 3 ≤ x < ∞.

orld
3 38 18

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,

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
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.

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

FX(2.5) = P(X ≤ 2.5) = P(X = 0,1,2) = 78.

2. FX has jumps at the values of xi ∈ X and the size of the jump at xi is equal to

P(X = xi).

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
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:

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, limx↓x0 F(x) = F(x0).
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,...,

A l
P(X = x) = (1 − p)x−1p. The It is cdf is

l JN
FX(x) = P(X ≤ x) =
 TU
∑x P(X = i) =

Wo
∑x (1 − p)i−1p = 1 − (1 − p)x.

rld
i=1 i=1easy to show that if 0 <p< 1, then FX(x) satisfies the conditions of Theorem 1.5.3.
First,
x→−∞lim FX(x)=0
4
since FX(x) = 0 for all x < 0, and

lim x→∞FX(x) = lim x→∞(1 − (1 − p)x)=1,

where x goes through only integer values when this limit is taken. To verify property (b),
we simply note that the sum contains more positive terms as x increases. Finally, to verify
(c), note that, for any x, FX(x + ε) = FX(x) if ε > 0 is sufficiently small. Hence,

limε↓0 FX(x + ε) = FX(x),

so FX(x) is right-continuous.
Example 1.5.5 (Continuous cdf)
An example of a continuous cdf (logistic distribution) is the function
FX(x) = 1
.
1 + e−x
It is easy to verify that
lim x→−∞FX(x) = 0 and lim x→∞FX(x)=1.

Differentiating FX(x) gives

ddxFX(x) = e−x
> 0,
(1 + e−x)2 showing that FX(x) is increasing. FX is not only right-continuous, but also

continuous.
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
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,
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-
 All
tributed if, for every set A ∈ B1, P(X ∈ A) = P(Y ∈ A). 5

JNTU

World
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
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
distributed. However, for no sample point do we have X(s) = Y (s).
Theorem 1.5.10 The following two statements are equivalent:
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).
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).

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). D

All JNTU
6

World
If is a discrete random variable with distinct values 1, 2,..., for which has positive proba bilities 1, 2,...,

, then the function defined as:

0 if , if is called the probability mass function of random variable .
Since 1(Probability of sure event is one.), we must have

ll
A Similar to PMF, probability tion defined as such that

J
1

density function is defined for continuous

 NTU W
random variable

o
. The funx

rld
1

The expected value of a discrete random variable is a weighted average of all possible values of the random variable, where
the weights are the probabilities associated with the corresponding values.
The expected value of a discrete random variable with proba bility
mass function (p.m.f) is given below:

The mathematical expression for computing the expected value of

a continuous random varriable with probability density function (p.d.f.) f(x) is, however, as follows:

Particular Case:

, which is defined as

0 for any
the th moment (about origin) of the probability distribution. Thus
(about origin) .
1(about origin)

A
l l
The the bility We The 1. 2. 3. 4. origin. consider function function Addition we Multiplication

all Expectation variables If Moment and the can and for expectations generalise is
J
are theorem: and discrete Generating called of independent if serves theorem: a

N
linear 1Moment distribution, , above 1 If exist. 2to

,..., Function Combination generate and If result and Generating 1 are and we are

T
as (m.g.f)of 2
any moments can random 2! are of generalise

Function. 1 random constants, Random a2 2! random variables, of 2

 U ...........
the variables, Variable: above probability 1then variable For For then

result Continuous Discrete , then Let 1 (about ! if distribution as all

..............
W
1, the 2,..., origin) expectations 1

1 of having the be any variate 1 exist.

o
the provided random

proba about , and if

rld
2

...........
!
..............
1
in
For Continuous For Discrete We see that the coefficient of ! gives
2
2(about origin)
Hence, Mean
2

Remar :

Similarly, we can show for continuous distribution. Thus th moment of about origin
0.
5. If , are two random variable s.t.

6.

0 then

!
..............

!
1(about origin) and

2
1

...........
2 2!
2
 , then
(about origin).
2

2
21
2

On differentiating above equation with respect to and then putting 0 , we get

A 0 In general, the moment generating function about the point

is defined as

...
where
, is the th moment about the point

1. being constant. 2. If 1, 2, 3,.................... are independent random variables,

ll
function of their sum 1 2 ....... is given by: 3. 1 2 ...... Effect of change of

origin and scale on m.g.f: changing both the origin and scale in as
JNT U
follows: 1 Let ustransform
2

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

W o
then to the the moment new generating variable by

rld
where is the m.g.f of about origin.
In particular
is nown as a standard variate. Thus the m.g.f of a standard variate is given by

Remar
E AMPLE 6.24
Let the random variable assume the value with the probability law:
1 ; 1,2,...
1
, where and are constants. M.G.F.of (about origin ) is given by :

If moment generating function exist it is unique.

About

1
...
1

then

22!
1

!. !

22!

2 ... !

1
 22! ...
2

2 ... !

Find the m.g.f of and hence its mean and variance.

Solution:

A 1
1

1 E AMPLE 6.25

ll
If the moments of variate 0 0.4, 1 0.6 Solution: The m.g.f. of variate X is:
JN T
are defined 2

U Wo
0 by 0.6; 1,2,3,... . Show that

rld
!
0.4 0.6
 0 1

on comparing we get the answer.

E AMPLE 6.26
Find the m.g.f. of the random variable whose moments are :
1 !2
Solution:
3 1
2
Hence, mean
0
1(about origin)
1
0 1
2(about origin)
1
1 and variance
1
1

2 1

2 1
1
1 2 ...

2
2
21
3

1 !2
!
 1 A 2 2

6.22 Define m.g.f of a r.v. Find m.g.f of (i) , (ii)

ll
6.23 Show that if is mean of random variables:

JNTU
World
!

A In statistics moment are defined vidual series or frequency

l l
distribution (i) about origin, origin (ii) is
 J N
mean, denoted or (iii) by any

T
r other as the point.

U
mean (discrete values and of continuous) powers of
Wo
the deviation in any indi ta en about three points:

and is defined as
In case of individual series x1,x2,...,xn, rth moment

rld
xr
r
n where r 1,2,3,4,...
If x1,x2,...,xn are the values of a variable x with

the corresponding frequencies f1, f2,..., fn respectively, then rth moment about origin is denoted by
xr f where r 1,2,3,4,...
r and is defined as
f
r

For grouped data, let x1,x2,...,xn be ta en as

the mid values then we have
f
r

xr f
where r 1,2,3,4,....
 In case of individual series x1,x2,...,xn, rth moment

about mean is denoted by A x r n where r 1,2,3,4,...

If x1,x2,...,xthe corresponding frequencies f1,

ll
f2,..., fn respectively, denoted by

JNTU
n are then the rth values
 W o
moment of a variable about mean x with is

rld
x r f
where r 1,2,3,4,...
For grouped data, let x1,x2,...,xn be ta en as
the mid values then we have
x r f
where r 1,2,3,4,...

In case of individual series x1,x2,...,xn, rth moment

about any arbitrary point a is denoted by
a r n where r 1,2,3,4,...
If x1,x2,...,xn are the values of a variable x with

the corresponding frequencies f 1, f2,..., fn respectively, then rth moment about any arbitrary point a is
denoted by
a r f
where r 1,2,3,4,...
r and is defined as
r and is defined as
r and is defined as
f
r
f
r

f
r

x
r
x
r

r and is defined as
x
x
x
For grouped data, let x1,x2,...,xn be ta en as
the mid values then we have
f
r
f
a r
where r 1,2,3,4,...

1. Second moment about a is equal to mean square deviation about a., i.e.
f
a 2 f
2. Second moment about mean is equal to variance, i.e.
2
3. First moment about origin is mean, i.e.
a fx
1
x
f

4. First moment about mean is always zero, i.e.

lJ
Al Calculate 5. 6.

The respectively.
 f If the series is symmetrical about mean, then all the central The value of a moment of order zero is
always 1. i.e.,

N
amount in the Find x x2 second poc ets 900 30 2 moment

1
T x2 73,800
6,400 of 80 n5 students about 12,100 110 origin. 5 in

U 14760 Rupees
14,400 college 120 were 40,000 200 Rs.30,80,110,120 x
 W 0
moments

Total 73,800
2

o
2 a

are zero.
fa 0

1
f

and 200
rld
first three moments about assumed mean 89 for the following table:
Mar s 81 87 89 90 91 94 96 No. of Students 7 11 15 8 4 3 2
x
x
1
f

2
x
x

a f
0

x 2 f
x
xfd x 89 fd fd2 fd3
81 7 8 56 448 3584 87 11 2 22 44 88 89 15 0 0 0 0 90 8 1 8 8 8 91 4 2 8 16 32 94 3 5 15 75 375 96 2 7 14 98 686
Total N 50 33 689 2571
rth moment about 89

l
Al It mean is Steps Compute better
 J
to the use n first this four method

N T
central when moments arithmetic

U
for the mean set is 2, an 4, 2571
 Wo 51.42
50 integer. The first moment about 6, 8.:

rld
x

is always zero.

1. Calculate mean x.
2. Obtain deviation x
x or d.
3. Obtain x
x 4,........
4. Obtain rth moment by
n x 2
or d
nr
,r 1,2,3,4,.......

In case of discrete and grouped series, the corresponding formula is

x
1

89 r,r 1,2,3..........

x 2, x
f x
1 89
 f x
2 89
f x
3 89

1 f x
r 89 N

f x
r

x
r

x 3, x
or fdr
x r f
,r 1,2,3,4........
f

1 fd
89 f N

1 fd2 689
89 2 f N
13.78
50

1 fd3
89 3 f N

33 50
6.6

xx x x
x 4
2 3 9 27 81 4 11 11611 1 1 8 3 9 27 81
20 0 20 0 164
Mean,
x 20 5
x n 4 Central moments

A
From the following frequency mean by direct method:
Size(x)
l l J N
12 16 20 x 4 8 Mean, N 2 2 1 4 1 f 10 16 12 64 fx 8 Frequency(f)

TU
4 8 4 d 0 20 8 fx 120 distribution, 428 compute 12 2 1

Wo
first four moments about the

16 20 4 1
rld
fd fd2 fd3 fd4
16 128 1024 8192 8 32 128 512 0 0 0 0 16 64 256 1024 8 64 512 4096
fd 0 fd2 288 fd3 384 fd4 13824
fx
x N
120

12
10 The first four moments about mean are:
0, always
1

x
2

x
3

x
4

20 5
x 2n 4
0 0
x 3n 4
164 41
x 4n 4
x 2 x
x 3 x

A
From the data given below compute method:
Mar s Frequency
 l l
Mar s Mid value,x f 0 10 5 1 10 20 15 2 20 30 25 3 30 40 35 4 10 Mean, The first four

J N
central moments are:

T fx
the 0 10 1 fx 5 30 75 x 140 250 N first four moments 10 20 20 30 2 3 10 10 20 d 0
U 250 25
40 20 20 fd 0 0 fd400 200 0 400 1000 10

2 about
W
30 40 4 8000 fd 2000 0 4000 6000 the mean 3 fd160000 20000 0 40000

220000

orld
by direct

d 220000
4 N4
22000
10
 Steps:
1. Ta e some convenient values as assumed mean, a,say.
0 0
1 10
fd2
2

288
N
28.8;
10
fd3
3

N
fd4
4

13824
N
1382.4
10
fd 0 0;
1 N 10
fd2
2

1000
N
100
10
fd3
3

384 10

6000 10
38.4
600

2. Obtain deviations from the assumed mean a i.e., calculate d x xa

3. Obtain fdx 2, fd 3, fd 4,............. and their totals
x x
fd
x, fdx2, fd 3, fd 4,.............
x x

4. Calculate v1 N fdx
,v fd
2 N x

2,v3 N
fd
x

3,v4 N
fd 45.
x Finally calculate central moments:

A
The first four moments about 4 in a distribution tively.Calculate moments based on mean .

1 f x
ll J
Given: v1 N Moments about arithmetic

NT U 4
mean: are 4 4 108

W
-1.5,17,-30 and 108 respec
orld
15.1875 142.3125
Steps:
1. Ta e some convenient value as assumed, say a.
2. Obtain deviations from the assumed mean a i.e, calculate d x x
a
3. Divide dx by a suitable value, say h, i.e.,
x
ds dxh
1 0,
2 v2
1 0
2 v2
3 v3
v
4 4

3v2v1 2v31
3v41 108
1.5 4 108
,
v21
30
17
v21
4v3v1 6v2v21
3 v3
4
180 229.5
3 17
1.5

1 f x
v2 N

4 2 17
f x
v3 N1
30
f x
v4 N1
30

1.5 2

3v2v1 2v31,
1.5 6 17
1.5 2 17
4 3
1.5 3 39.75
ah
4 v4
2.5 14.75
1.5 2
3

4v3v1 6v2v21
3v41
4. Calculate
v1 fds

1 fd
f N s

v2 fds

2 1 fd 2v 1 fd 1 fd
f N s 3 fds 3 f N s
3v
4 fds 4 f N s
45. Finally calculate central
moments:
1 0, always

A
Calculate first four central moments from the following data:
 ll
Class Frequency Class Frequency (f) 0 10 10 20 20 30 Here 30 40 40 50 a

J N T
Total 25,, h 10 0 10 10 20 1 3 20 1 3 5 7 4 Mid value 15 25 5

and

U 10
(x) ds 20 30 5 x 10 25 30 40 7 fds 2 2 1 3 0 0 35 1 7 45 2 8 10 v1 fds f 20

0.5

fd
v2 s
2 30 1.5
f 20
fd
v3 s

3 28 1.4
f 20

W
40 50

orld
4 fds 24 3 0 7 16 30 fds 3 8 3 0 7 32 28

fd 416 3 0 7 64
s

90
v4 fds
4 90 4.5
f 20

2 v2
3 v3
4 v4
v21

3v2v1 2v31

4v3v1 6v2v21
h2
h3
3v41
h4
104 37625
Charlier has given the following formula which can be applied the following totals while calculating
moments.

fd
x,
fd 2,
x
fd 3,
x
fd
x
4or fds, fds2
, fd 3,
s
fd
s
Al
4 Second Fourth

l J
These Third The Find First Step central the chec s moments
 N
f dmoments are x 1 applied f dabout x 4 of distribution origin. Charlierss stepwise:

T U W
Chec whose formula mean is 5 are for testing

o
the accuracy of Up to Moment

0,3,0 and 26 respectively.

rld
6
f d 1 x 1 3f dfd x 4x 1 2 4 fd x3 3 fdx 2fdx 3fdx f 2 fdx 2
fd 4 fd fd
fd x x 2 x ff x f First Second Third Fourth

If are:
4 are first four moments about origin then first four moments about mean
Hence central moments are:
0
1

1,
2,
3,
2 v2
v
3 3

v
4 4

h2
1.5
h3
1.4
h4
4.5
v21

3v2v1 2v31
4v3v1 6v2v21
0.5 2
3 1.5 0.5 2 0.5 3
4 1.4 .5 6 1.5 0.5 2
1 x
2
3
 4
2
3 3
4 4
102 125
2 or
1
1
1

3v41
2
3 6
2 x2
3
1
2
1
103
3 0.5 4
2
600
4
1

= Problem 1 The probability density function of a continuous random variable X by f ( x ) = Ke- x .

Find K and C.D.F of X Solution: Since it is a probability density function,

A l
-∞ ∞ f ( xdx ) = 1 Problem variance Solution: V Problem Solution: F ( 3 X+
 l
= 3 A ( x ) = 3 x 2 ; 0 ∫-∞ ∞ Ke - x dx
=
4 of 2 Y 3 ) X X = 1 ∞2

∫
Ke - x dx = 1 0
0

J
and + 4 Y ∫ Y.

N TU
+ ∫ are independent = + - random variables 9 Var (
 X ) + 16 Var ( Y ) + 24 Cov ( XY ) 9 × 2 + 16 × 3 + 0 ( ∴ X & Y are independent 18 48 66.
Continuous random variable X ≤ x ≤ 1. Find ‘a’ such that P ( x ≤ a ) = = has P ( x

W o
- with cov( a probability > a ) variance 2 and 3. XY ) = 0 ) density is

r
given Find the

function

ld
2K
K=1 .
⌈ │ ⌊ e - - 1x ⌉ │ ⌋ ∞ = 1 2 K = 1 2 Therefore,
1
f ( x ) = 2e x, -∞ < x

<0=12
e-
x

,0<x<∞

1 1
For x ≤ 0, the C.D.F is F ( x ) = x ∫ -∞ We know that the total probability =1 2 e x dx = 2

e x For x > 0, F ( x )
0
-∞

1 2 edx x x

e - x dx 0
1 2 1 2 1 2 (1 e - x ) 1 2

(2 e -
x )
Given P ( X ≤ a ) = P ( X > a ) =
K ( say ) Then K + K =
1 1 P X≤a =1
1 K = 12 i.e., P ( X ≤ a ) = 2 & P ( X > a ) = 2 Consider ( ) 2 i.e., ( )

l
0
Al ∫ =a ∫

f x dx = 12
 ∫ a

3x2

JNTU
dx = 1 2 ⎛ │ ⎝

World
∫ +=∫ +3x 3a 3 0⎞ │ ⎠

⎛
= 3a = │ ⎝ 1 2
a=
⎞
12 11/3 2 │ ⎠ . Problem 4 A random variable X has the p.d.f f ( x ) given by ( )=⎧⎨⎩
; 0
0- ;> ≤ 0 Find the value of C and cumulative density function of X . Solution:
Since f ( x ) dx

∫
1 f x Cxe x if x if x ∞ = -∞
∞ ∫
Cxe - x dx = 1 0 C ⌈ ⌊ x ( -e ) -( e
-
x

-
x
) ⌉⌋ ∞0 =1C=1∴ f( x) =⎧⎨⎩
; 0
xe 0 - x x > ; x

≤ 0 CDFF . . ( x ) = ∫ x f ( xdt ) x = ∫

te - t dt = ⌈ ⌊ - te - t 0 0

- e - t ⌉ ⌋ 0 x = - xe - x - e - x + 1
= 1 - ( 1 + x ) e- x for x ≥ 0.

Problem 5 If a random variable X has the p.d.f f ( x

) = ⎧ │ ⎨│⎩ 1 2
( )
x + 1; - 1 < x < 1 0 ; otherwise
, find the
mean and variance of X . Solution:

Mean= 1 xf ( x ) dx ( )( ) -1 1 1 2 - 1 x x 1 dx 1 1 2 -
1

x2
x dx
1 = ⎛ │ ⎝ 3+ 2⎞ │ ⎠
1 =( )( )
1xx
1232-
3
 A ll J
2 ′ = 1∫ 2e 11 Problem variance is 8 4 ⇒μ x f -

N -λ
λ = xdx = 1 2 - ∫ 113 +2

T
e - λ λ + ⇒⇒ λ λ 4 2 + + 3 4 λ
( )( 2 - λ x x dx = 1 2 4 2 - = 1 0 ) = 0 ⇒λ 2 = 1 ( or ) λ 2

= - 4 λ =Variance 1= ( λ 2
Comment the following: “The ⌈ │ ⌊ 4 x 4 + x 3 3

⌉│⌋ -1 =12⌈│⌊ 14+13-14+13

 Variance μ ′
⌉│⌋ == 1 2 2 .2 3 -

( =μ 1 3 ′ ) = 1 3 - 1 9 = 3 9 - 1 = 9 2
1 2

U
. Problem 6 A random variable X has density function given

Wo
by ()

rld
=⎧⎨⎩
e - λ λ cannot be negative)
mean of a binomial distribution is 3 and
Solution: In a binomial distribution, mean (np)>variance (npq). - ≥ < . Find the moment generating
function. Solution:
()( )() 2

00

f x 2 e 2x; x 0
0;x
0

M t E e tx e tx f x dx
∞∞ X = =∫ =∫
e tx e x
2 -

dx

2
= ∞ ∫
e ( t - )2
x
dx 0= 2 ⌈ │ ⌊ e ( t - )2

x∞ t-20
⌉ │⌋

=2
2-t

,t<2.
Problem 7 If X is a Poisson variate such that P ( X = 2 ) = 9 P ( X = 4 ) + 90 P ( X = 6 ) , find the

variance. Solution:Given P ( X = 2 ) = 9 P ( X = 4 ) + 90 P ( X =

6) 246 2! 9 4! 90 6!
Since variance = 4 & mean = 3 , we have variance<mean. Therefore, the given statement is wrong.
Problem 9 If X andY are independent binomial variates
⎛ ⎞ ⎛ 1 ⎞
B │ ⎝ 5, 1 2 │ ⎠ and B │ ⎝ 7, 2 │ ⎠find P [ X Y+ = ]3 Solution:X Y+ is also a binomial
⎛ 1 ⎞
variate with parameters n 1 + n 2 = 12 & p = 1 2 ∴ P [ X + Y = 3 ] = 12 C 3 │ ⎝ X 2 │ ⎠ 3 9
 All JNTU

W
Z = - = 0.5 - P [ 0 ≤ Z ≤

orld ⎛ 1 ⎞ =
0.17 ] = 0.5 - 0.0675 = 0.4325 │ ⎝ 2 │ ⎠ 2

55 4 , find P [ X > ]0
10 Problem 10 If X is uniformly distributed with Mean1and Variance 3
Solution:
If X is uniformly distributed over( ,a b ) , then

E ( X ) = b + 2 a andV ( X ) =
(
b 12

)2
-a

∴ b+2
a = 1 ⇒a + b = 2 ⇒( b 12 - a ) 2

=43

⇒( b - a )

2 = 16 ⇒a + b = 2 & b - a = 4 We get b = 3, a = - 1 ∴ a = - 1& b = 3 and probability density function

of x is 1f ( x )

= ⎧ │ ⎨│⎩0;

4; Otherwise - 1 < x < 3 P [ x > 0 ] = ∫ 0

3 1 4 dx = 1 4 [ x ] 30

P⌈ Y 3 ⌉
= 3 4 . Problem 11 If X is N ( 2,3 ) ind │⌊ ≥ 2 │ ⌋ where Y + 1 = X . Solution:
⌈ ⌉ ⌈ ⌉
P │⌊ Y≥32 │⌋ =P │⌊ X-1≥32 │⌋

= P [ X ≥ 2.5 ] = P [ Z ≥ 0.17 ] , where 3

2

1 that a man th
Problem will hit the 12 target If the for probability the first time is 4 in the 7 trial?
will hit a target, what is the chance that he

Solution:The required probability is P [ FFFFFFS ] =

⎛ ⎞ ⎛
P ( F ) P ( F ) P ( F ) P ( F ) P ( F ) P ( F ) P ( S ) = q 6p = │ ⎝ 3 4 │ ⎠ 6. │ ⎝ 1 4
⎞
│ ⎠ = 0.0445 . Here p =probability of hitting target and q = 1 - p .
Problem 13 A random variable X has the following probability function: Values of
:01234567
: 0 2 2 3 22 27
2
70 1, x XP X K K K K K K K K Find (i) K , (ii) Evaluate P ( X < 6 ) , P ( X ≥ 6 ) and P Solution:
(i)

Al l
K (ii) + 2 As K P (iii). ( Now Now + X P 2 ( K < x Determine P 6

J
) + P ) Since cannot ( K 3 ( 10K0 = = K X = < P 10 1 + ≥ 10 ( X 21 K +9K−1=0 X 6

∑ P
N
+ be = < ) 2 10 the = 2 = = or 5 + negative 0 1 ) 1 2 + ( ) - = distribution - x K
 =
T
10 + K ) 2 100 P P 81 2 P ( ( + + = ( X X 7 10 - X = 3 K 1 < = K 100 = 19 + 2 16
=

U
1 ) + 100 ) ) + 1 10 function K + 1P ... + = ( + X 1 ... P = = ( 100 2 of X 81 )

+X=P5.) ( = K + 2 K + 2 K + 3 K = 8K = 10 8= 5 4.

W orld
( 0<X <5) X=3) +P( X=4)

(iii) The distribution of X is given by F X( x ) defined by

( F Xx ) = P ( X ≤
 X
x)
:01234567
FX x

: 0 1 10 3 5 10 10 4 81 83 5 100 100 1 Problem 14 (i). If the probability distribution of X is

given as
X: 1 2 3 4 P X

l
: 0.4 0.3 0.2 0.1 Find P ( 1/2 < X < 7/2 X > 1 ) (ii). find Solution: (i) A If P ( x ) =

⎧ │ ⎨│⎩
; 1,2,3,4,5
150 x x = ; elsewhere (a) P { X = 1 or 2 } and (b) P { 1/2 < X < 5/2 x > 1 } P { 1/2 <

{(
l
X < 7/2/ X > 1 } = P 1/2 < P X (
 JNTU W = =
1 - 2/15 1/15

= =
14/15 2/15 14 2 1

.
orld
7 <X

) }
7/2 ⋂X>1

>

1
) = P ( X = 2 or
3
)P(X=
2,3 or
4
) =P( XP( X=2) +P( X
=3
) =2) +P( X=3) +P( X
=
4
) = 0.3 0.3 + 0.2 + 0.2 +
0.1
=
0.6 0.5
=5 .
6
(ii) (a) P ( X = 1 or 2 ) = P ( X = 1 ) + P ( X =
= + = =
2) 15 1 15 2 15 3

1 5 (b) P ⎛ │ ⎝ 1 2 < X < 5 2 / x > 1 ⎞ │ ⎠

(
=P⎧⎨⎩⎛│⎝12<X<52( ⎞│⎠⋂ X

)
>1 ⎫⎬⎭PX

>1
) = P { ( X = 1 or 2 ) ⋂ ( X
>1
) } P( X
>
1
) = 1 - P P ( X( X = = 2
)1
)
()
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 3

K 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 .
Solution: a) Since ∑
P ( X ) = 1 0.1 K 0.2 2K 0.3 3K
 K= =
16K 0.6 16K 0.4 0.4 6 15 1 b) P ( X < 2 ) = P ( X = - 2, -
= + + +
1, 0 or 1 ) = P ( X = - 2 ) + P ( X = - 1 ) + P ( X = 0 ) + P ( X = 1 ) 10 1 15 1 15

A ll
15 2 3 2 6 4 15 1 30 30 2 P ( - 2 < X c) The distribution d) Mean EF

J
of X( X = < 2 ( x ) ) =F====+PX X
P 15 1 P ( 1 + x ( X function ( + 15 X
+

+
N
X3≤+15=+=::x2 - - ) 15 = 1,0 10 1 2 = 1 ) 15 2 of 6 + or P = X=1( 25
 T U
1 ) X is = given 0 0 ) + 1 P by ( X 2 F X= ( 3 1 x ) )

W
defined by

X is defined by E ( X ) =∑ xP ( x ) ) = ⎛ │ ⎝ - 2 × 10 1 ⎞ │ ⎠ + ⎛ │ ⎝ - 1 × 15 1 ⎞ │ ⎠ + ⎛ │ ⎝
1 ⎞ ⎛ 2⎞ ⎛ 3⎞ ⎛ 1 ⎞ =- -
0× 5 │ ⎠ + │ ⎝ 1 × 15 │ ⎠ + │ ⎝ 2 × 10 │⎠+ │⎝3× 5 │⎠ 1 5 15
+ + +
1 15 2 35 3

= 16 .
orld
5 15 1 6 11 1 30 2 4 5 1
Problem 16 X is a continuous random variable with pdf given by
( ) F X = ⎧│ │ ⎨ ││⎩
Kx in x
6 02 KK 0≤

2 in 2 x 4 - Kx in 4 x

6 elsewhere ≤ ≤ ≤ ≤ ≤ Find the value of K and also the cdf F X( x ) . Solution:

∫
Since ∞ F ( x ) dx

=1∞ ∫ 2 Kxdx + ∫ 4 2 Kdx + ∫ 6

( 6 k - kx ) dx = 1 0 2 4 K ⌈ │ │ ⌊ ⎛ │ ⎝ x 2 2⎞ │ ⎠ 02+ ( 2 x ) 42

+∫

64 ⎛│⎝6x-x2

l
Al 2 ⎞ │ ⎠ 64⌉ │ │ ⌋

JNTU
= ∫ +∫ +∫
World
= K ⌈ ⌊ 2 + 8 - 4 + 36 - 18 - 24 + 8

⌉ ⌋ = 1 8 K =1 K = 18 We know that ( ) ( ) 1 x F Xx = ∫
f x dx

F
-∞ If x < 0 , then X ( x ) = x∫
f x dx
( )
=0

-∞ If x∈ ( 0,2 ) , then F X( x ) = x ∫
f x
( )
dx

F
-∞ X ( x) =0

∫ f ( x ) dx + ∫ x

f x
( )
dx

-∞ 00 = ∫ -∞ x 0 0 dx + ∫ Kxdx = ∫ 0 -∞

0 dx + 1 8

∫ 0x
xdx

= ⎛ │ ⎝ x 2 16 ⎞ │ ⎠ 0 x = 16 x 2 ,0 ≤ x ≤ 2 If x∈ ( 2,4 ) , then F X( x ) = x

∫
f x
( )
dx

F
-∞ X ( x ) = 0 ∫ f ( x ) dx + ∫ 2f ( x ) dx + ∫ x

f x
( )
dx
-∞ 0 20 2x

0 dx Kxdx 2
Kdx
-∞

02

22 = ∫ 0 8 x dx + ∫ 2x 1

4 dx = ⎛ │ ⎝ 16 x 2

⎞│⎠0

+ ⎛ │ ⎝ 4 x ⎞ │ ⎠ 2 x = 1 4 + 4 x -1
2 = 4 x - 16 4 = x 4

- 1 , 2 ≤ x < 4 If x∈ ( 4,6 ) , then F X( x ) = 0 ∫ 0 dx + ∫ 2 Kxdx + ∫ 4

Kdx
2 + ∫ x

k
(6
x
- )
dx
-∞
024
 ll JNTU
() A
World
∫ = =∫ 2 0 8 x dx + ∫ 42 1 4 dx + ∫ 4

x 1 8 6 - x dx = ⎛ │ ⎝ 16 x 2 ⎞ │ ⎠ 0 2 + ⎛ │ ⎝ 4 x ⎞ │ ⎠ 4 2 + ⎛ │ ⎝ 6

8 x - 16 x 2

⎞ │ ⎠ 4 x = 1 4 + 1 - 1 2 + 6 8 x - 16
x 2- 3 + 1 =
+ - + x-x - +
4 16 8 12 16 2 48 16 = - x 2 + 12 16

F
x - 20 ,4 ≤ x ≤ 6 If x > 6 , then ( x ) = 0 ∫ 0 dx + ∫ 2 Kxdx + ∫
X 4 2
Kdx
+ ∫ 6

k
( 6-x)
dx
+ ∫ x

0
dx
-∞

0246 = 1, x ≥ 6 ∴ F ( x ) = ⎧││ │││ ⎨│-│ ││ │⎩ 0 x 161 4( x 1 )

16 1 ( 20 12 x 1 ; x
≤0
2

;0 ≤ x ≤ 2 X

-;2≤x

≤ 4 - + x 2)

;4 ≤ x ≤ 6 ; x ≥ 6
Problem 17 A random variable X has density function f ( x ) = 1
,
K 2 0 , ⎧ │ ⎨│⎩

+x

-∞ < x < ∞ Otherwise Determine K and the distribution function. Evaluate the probability P ( x ≥ )0
. Solution:
Since ∞f ( xdx
) 1 -∞
∞-∞ ∫
1
1
+ K x 2 dx = K ∞∞
∫
1

ll
1
+ dx x 2 = K ( tan - 1 x ) ∞ -∞
= ⎛
1K │⎝ A ≤ ≤
π ⎛
2- │⎝-
π ⎞
2 │
 J ⎞
⎠ =∫ - │ ⎠ = 1 Kπ = 1 K = π 1 F X ( x ) = x ∫ -∞ f ( x ) dx = x ∫-∞

K =
1+ x 2 dx

π 1 ( tan

N T U π
-1 x ) -∞ x =π1 ⌈ │ ⌊ tan -1 x-⎛│⎝ -

World
2⎞│⎠
= ∫ -=
-
-

- = ⌈ ⌊ -- -⌉ ⌋ ⌉ │ ⌋ = π
π
1⌈│⌊

2 + tan - 1 x ⌉ │ ⌋ , -∞ < x < ∞ P ( X ≥ 0 ) = π 1 ∞ ∫

0

+ dx x 2 = π
1 ( tan -
1
x) ∞0 =π1π

2 tan 1 0 1 2 ⎛ │ ⎝ - - ⎞ │ ⎠ = .
Problem 18 If X has the probability density function ( )
= ⎧ ⎨⎩ - > find K , P [ 0.5 ≤ X ≤ 1 ] and the mean of X .

Solution:Since f ( x ) dx

1 f x Ke 3 x , x 0
0 ,otherwise

∞ ∫= ∫ -∞ ∞

Ke - 3
x
dx = 1 0

K
⌈ │ ⌊ e - -3

3x ⌉ │ ⌋ ∞0

= 1 K 3= 1 K = 3 P ( 0.5 X 1 ) 1 f ( x ) dx 3 1 e 3 x dx 0.5 0.5

e e
3 3 1.5

3
e 1.5 e
3

∞ Mean of X = E ( x ) = ∫ xf ( x ) dx = 3 ∫∞
xe - 3
x
dx
0 03 3

ll J
0 B=3⌈ │⌊ A X Rs W Rs Bx⎛│⎝-e3-

NT U
W B W Rs Rs x ⎞ │ ⎠ ∞

World
- 1 ⎛ │ ⎝ e -9

x ⎞│⎠⌉ │⌋
=39×1=
13

Problem 19 A random variable X has the P.d.f f ( x ) = ⎧ ⎨⎩

2 x ,0
0 ,Otherwise
x1 P⎛ X 1 ⎞ P⎛ 1 x 1
< < Find (i) │⎝ < 2 │ ⎠ (ii) │⎝ 4< <
⎞ ⎛ 3 1 ⎞
2 │ ⎠ (iii) P │ ⎝ X > 4 / X > 2 │ ⎠ Solution:

(i) P ⎛ │ ⎝ x < 1 2 1/2 ( )

0 ⎞│⎠= ∫ f 1/2 x dx = ∫
0 1/2 2 xdx = 2 ⎛ │ ⎝ x 2 2 ⎞ │ ⎠ 0

=28×1=14
(ii) P ⎛ │ ⎝ 1 4 < x < 1 2 ( )
1/2 ⎞│⎠= 1/2 1/4 ∫ f x dx = 1/2 ∫
1/4

2 xdx = 2 ⎛ │ ⎝ x 2

2⎞ │ ⎠ 1/4 = 2 ⎛ │ ⎝ 1 8 - 32 1 ⎞ │ ⎠ = ⎛ │ ⎝ 1 4 - 16 1 ⎞ │ ⎠ = 16
3.
(iii)
⎛313⎛│⎝>34/>12

⎞│⎠=│⎝>4⋂>2⎛│⎝>124⎞│⎠⎞│⎠=⎛│⎝⎛│⎝>⎞│⎠>12⎞│⎠
()
PXXPXPXX
PXPX

P ⎛ │ ⎝ X > 3 4 ⎞ │ ⎠ = 3/4 ∫ 1 f 1 x dx = ∫
3/4 2 xdx = 2 ⎛ │ ⎝ 1 x 2 2
⎞ │ ⎠ 3/4

= 1 - 16 9 = 16 7 P ⎛ │ ⎝ X > 1 2 ⎞ │ ⎠ = 1/2 ∫ 1 f ( x )

dx = 1 ∫
1/2 2 xdx = 2 ⎛ │ ⎝ x 2
1 2 ⎞ │ ⎠ 1/2
⎛ ⎞
=1-14=347P │⎝X>34/X>12 │⎠

= 16 3 = 16 7 × 4 3

= 12 7 . 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.
Solution:
Let X denotes the amount that he expects to receive
W B 5 7 3 1 &2 2 &1 3
15 20 25 30
7×6×
5 P ( X = 15 ) = P ( 3 Black balls ) = 12 7 C C 33

= 12 1 × × 11 2 × ×

3 10 = 44 7 1 × 2 × 3 P ( X = 20 ) = P ( 2 BW 1 ) = 7 C 12 2 .5 C 3
1

All J N T =
= + +
 U World
320 10 E(x)

7 2.5 2 14
= Rs .32 / - C = 44 21 P ( x = 25 ) = P ( 2 W 1 B ) = c 12 c 3 c = 44 P ( X = 30 ) = P
( 3 W ) = 12 5 C C 33
=

44 2 E ( X ) =∑
=
xP ( x ) = 15 × 44 7 + 20 × 44 21 + 25 × 14 44 + 30 × 44 2 935
= Rs .21.25
44 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: Let X denotes the amount he receives
R B3 2

2B1R1B2
RX Rs 20 Rs 30 Rs 40

()() 22

2 ×1 P X = 20 = P 2 Black balls = 2 5 C

C = 1 5 × ×2 4 = 10 1 1 × 2 P ( X = 30 ) = P ( 1Re d &1 Black ball ) = 3 C 1 5 × C2 2

C1

10 6 P ( X = 40 ) = P ( 2 Re d balls ) = 5 3 C C 22
=
1 6
10 3 E ( x ) =∑ xP ( x ) = 20 × 10 + 30 × 10 + 40 ×
 3
10 20 180 120
10
Problem 22 The elementary probability law of a continuous random variable is
f ( x ) = ye0 - b ( x - a )
,
a ≤ x < ∞ , b > 0 where a, b and y 0are constants. Find y 0, the rth moment about the point x = a and
also find the mean and variance.
Solution: Since the total probability is unity,
f ( x ) dx 1 = Γ + =
In particular r =

μ ′ 1μ ′ 2
1 1 =b 2 =b 2 ∫
Mean 1 ∞ = -∞
y0

∞ ∫
e -b( x-a)
dx = 1 0
0

ll
A Variance Problem =a
=
= + 23 b
μ 2
μ 22
′′-
The - = b ( 1a
 J μ ′
2 +1
=
first )1 b b 2 1
2 ⌈ │ ⌊ y e -b( x-a) ∞- b 0

NTU

World
.
four moments of a distribution about x = 4 are 1,4,10 and 45 respectively. Show that the mean is 5,
variance is 3, μ 3 = 0 and μ 4 = 26 . ⌉ │ ⌋ = 1
 1 μ′ ( rth
y 0 ⎛ │ ⎝ b ⎞ │ ⎠= 1 y 0 = b . r moment about the point x = a ) = ∞ ∫
( x - a ) rf ( x )
dx

b
-∞ = ∞ ∫
( x - a ) re -b( x-a
) dx
a Put x - a = t , dx = dt , when x = a , t = 0 ; x = ∞ , t = ∞

∞ =b∫

t r e - bt dt
r1
0 b( ) b ( r+
r!
)1 br
′ ′ ′ ′
Solution: Given μ 1 = 1, μ 2 = 4, μ 3 = 10, μ 4 =
μ ′ r
45 r = th

μ ′μ ′
moment about to value x = 4 Here A = 4 Hence mean = A Variance = 2 +=μ1 2 =-

4+1= ( μ ′) 5=4-μ
1
2
3 =
μ ′
3 1-=
3μ
3 2.
′μ ′
1 +

2
( μ ′ ) = 10 - 3 ( 4 )( 1 ) + 2 ( 1 ) =0 μ
1 3
3
4 =
μ ′ 4μ ′μ ′ 6μ ′
4 - 3 1 + 2 ( μ ′)1 2 -

3
( μ ′) 1

μ = 26 .
= 45 - 4 ( 10 )( 1 ) + 6 ( 4 )( 1 ) 2 - 3 ( 1 ) 4
4

Problem 24 A continuous rth moment of X about Solution:

 All J
Since 2 =∞ ∫ 0 random the origin. Hence Kx 2

N TU +
e - x dx = 1 0 variable find X mean has ∞ ∫ -+- = the

W
p.d.f f ( x ) = kxe 2 and variance of X.

′ ′
Putting r = 1 , μ 1 = 3! 2 = 3 r = 2 , μ 2 = 4! 2 =
 or
12 -x
, x ≥ 0. Find the

ld
′
K ⌈ │ ⌊ ∴ Mean = μ 1 = 3 x ⎛ │ ⎝ ∞ e - x - 1 ⎞ │ ⎠ - 2 x ⎛ │ ⎝ e - 1 ⎞ │ ⎠ x + 2 ⎛ │ ⎝ e - -

1
x⎞ │ ⎠ ⌉ │ ⌋ = 1 2 K = 1 K = 12 .
μ

r ′ ∞ =∫
x
r

f x dx
( ) 0 = 12

∞ ∫ 0

+2

-( 3) 1 ()0
x r e x dx
12 e rx x dx
r
2
2!
μ ′
Variance = 2 -

( μ ′) 1
2

μ = 12 - 3
i.e., 2 ( )2
= 12 - 9 =3
Problem 25 Find the moment generating function of the random variable X, with

ll
probability density function ( ) ∞0 f x = A
JNT ⎧
⌈ │ ⌊ - -⌉ │ ⌋ │ ⎨│⎩
 x
U World
20 -x =k=1

for ≤ x <
1 for 1 ≤ x
′ ′
< 2 otherwise. Also find μ 1 , μ 2 . Solution:

M
X ( t ) = ∞∫
e tx

f xdx
( ) -∞ = ∫ 1 e tx xdx + ∫ 2

e tx ( 2 - x ) dx 0 1 = ⎛ │ ⎝ xe tx t - e t tx 2 ⎞ │ ⎠ 1 0 + ⌈ │ ⌊ ( 2-x)
e t tx - ( - 1)

et
2

⌉│⌋ 12

= - + 1 +
e tt e tt 2 t 2 e2
t - -
t 2 e tt

e t t 2 e t1 2 t tx = ⎛ │ ⎝ - ⎞ │ ⎠ = ⌈ │ ⌊ 1 + 1! t + t 2! 2 + t 3! 3

+ ... - 1 ⌉ │ ⌋ 2 = ⌈ │ ⌊ 1 + 2! t + t 3! 2 + t 4!
3

′ . t 1
+ ... ⌉ │ ⌋ 2 μ 1 = coeff of 1!= μ 2 ′ = coeff . of t 2

2! =
7
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.

Solution:Total probability=1

∞ ∴ ∫
ke - x dx = 1 0

e x 10
1

M
∞ X ( t ) = E ⌈ ⌊ e tx ⌉ ⌋ = ∫ e tx e - x

dx
= ∞∫
e( )1
t-

dx
x

l JN
00 =⌈ │⌊ ( )1
1
0 = et - ∞ t -
Al = ∫
TU World
= ∫

x ⌉ │⌋ =1-1t,t<1=( 1-t) -
1
= 1 + t + t 2 + ... + t r + ... ∞
′ t
μ r = coeff . of r 2
′ ′ ′ ′
! = r ! When r = 1 , μ 1 = 1! = 1 r = 2 , μ 2 = 2! = 2 Variance = μ 2 - μ 1 = 2 - 1 = 1 ∴ Standard
deviation = 1.
Problem 27 Find the moment generating function for the distribution whose p.d.f is
f ( x ) = λ e - λ x , x > 0 and hence find its mean and variance.
Solution:

M
X ( t ) = E ( e tx ) = ∞ ∫ e tx f ( x )

dx
= ∞∫
λ
e λ e dx
- x tx 00

=λ

∞ ∫
e - x ( λ - t ) dx 0

=λ⌈ │⌊ -e
( -λ
x( λ

- -tt )) ⌉ │ ⌋ ∞0
= λ λ-
t

Mean = μ 1 ′ = ⌈ │ ⌊ dt d M ( t ) ⌉ │ ⌋ =0 =⌈││⌊ ( λ - λ t) 2

⌉││⌋
=

()
0 =( )
()
1X

tt λ μ 2′ = ⌈ │ ⌊ d 2
dt 2 M X

t⌉ │⌋ t=0 =⌈││⌊ λ2λ-t

3 ⌉ │ │ ⌋ Variance = μ 2 ′ - ( μ ′)
1
2

2 1
= λ 2- λ 2 =
1
λ 2 =λ22
t

- Problem 28 Let the random variable X have the p.d.f ( )

= ⎧ │ ⎨│⎩ > Find the moment generating function, mean & variance of X .
Solution:
()( )() /2

0
x

fx
1
2e 2 , x 0 0 , otherwise .
M t E e e f x dx
X tx ∞ tx -∞

e
tx 12
e
-

x
dx
∞

∞⎛│⎝-⎞│⎠ =12∫ 0

⌈ ⎛ │ ⎝ - ⎞ │ ⎠ = │ │ │ │ ⌊ ⎛ │ ⎝ 1 - ⎞ │ ⎠ e t 1 2 x dx 1 2 ⌉ │ │ │ │ ⌋

e t2
x

12

=-0

1 < 1 2 t , if t 1 2 . E ( X ) = ⌈ │ ⌊ dt d M X

(t)⌉ │⌋ t=0 () 2

Al l
0 ⌈=⌈││⌊ Variance Mean E ( = X = np 2 =│⌊ ) ==E

J N
;Variance n n ( 2 X ( - p n 2 2 - ) + - 1 np ) ⌈ ⌊ p + E = ( 2 1 [ npq + X -
 np ] p ⌉ ⌋ ) 2 2 1 - 2 t ⌉ │ │ ⌋ t

= ( ) =2EX 2

= ⌈ │ ⌊ dt d 2 2

MX

(t)⌉ │⌋ t=0 ⌈( )

TU World
+ + ⌉ │ ⌋ = n 2 p 2 + npq = npq b) Let X : the number of times the dice shown 5 or 6 = │ │ ⌊ - ⌉ │
│⌋ =( )( )()
812

t 3t
=

0 8 Var X = E X 2 - ⌈ ⌊ E X ⌉ ⌋ 2 = 8 - 4 = 4 . 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: a) A random variable X is said to follow binomial distribution if it
P X x
assumes only non- negative values and its probability mass function is given by ( = ) = nC x p
x q n-x

x n and q p
, = 0,1,2,..., =1- . M.G.F of Binomial Distribution about origin is
M
X
( t ) = E ⌈ ⌊ e tx ⌉ ⌋ = ∑
n e
x =0 tx

P X=x
(
)
=∑

n nCxPq
x

-
xn

e =∑ nC
x tx x = 0 n x

() 0

pe q -
x t n

M t
x= X ( )
=
( q+
pe
) Mean of Binomial distribution
t n

Mean = E ( X ) =
′ n
MX ( )0 = ⌈ │ ⌊ ( q + pe ) t
n-
1

pe
t

⌉│⌋ t=

np
=

Since q + p = 1 E ( X ) =M ′′ ( 0 ) n ( n 1
2
X

q pe
)( t ) ( pe )
n-2 t 2
npe
t ( q pe t

)
n-
1
t=
0

[ P 5 or 6 ] = 1 6 + 1 6 = 1 3 P∴ = 13 and
q=2
3 Here n =6 To evaluate the frequency of X ≥ 3
By Binomial theorem,

A ll J
[ ] 26P X = r = 6 C r b) Probability = E of getting x2

N
-⌈⌊EX⌉⌋

TU =1 .
one head ⎛ │ ⎝ = λ with one coin 2 13⎞│⎠r
World
⎛│⎝23

⎞ │ ⎠ -r
where r = 0,1,2...6 .
P[ X≥3] =P( 3) +P( 4) +P( 5) +
⎛ 1 ⎞ ⎛ 2 ⎞ ⎛ 1 ⎞ ⎛ 2 ⎞
P ( 6 ) = 6 C 3 │ ⎝ 3 │ ⎠ 3 │ ⎝ 3 │ ⎠ 3+ 6 C 4 │ ⎝ 3 │ ⎠ 4 │ ⎝ 3 │ ⎠ 2+ 6
⎛ 1 ⎞ ⎛ 2 ⎞ ⎛ 1 ⎞
C5 │⎝ 3 │⎠5 │⎝ 3 │⎠+6C6 │⎝ 3 │⎠6

= 0.3196 ∴ Expected number of times at least 3 dies to show 5 or 6 = N × P [ X ≥

]3 = 729 × 0.3196 = 233 .
Problem 30 a) Find the m.g.f. of the geometric distribution and hence find its mean and variance.
b) Six coins are tossed 6400 times. Using the Poisson distribution, what is the approximate
probability of getting six heads x times?
= M t = E e⌈ tx
Solution: a) M.G.F about origin X( ) ⌊
⌉
⌋ =∑

∞ e tx x = 0 λ x
=
e -λx ! () 0 ! e-
∑
λ

∞
λ
xx= e tx
M
= e - λ eλ e t X (t)=
eλ ( t- 1 )
e Mean and variance using M.G.F
Mean = E ( X ) =
 ′
MX ( )0 = ⌈ │ ⌊ e λ ( e - t

e
)λ t

⌉│⌋
t=

E
0 =λ (X 2

) =M X

0
′′ (
e λ( ) e λ(
)=⌈│⌊ ( λe )
t
2
et
-1
+ et
-

) e
1
λ t

⌉│⌋

t=0 = λ 2+ λ ∴ Variance ( )()

6 ∴ The probability of getting six heads with six coins ⎛ │ ⎝ 1 2 ⎞ │ ⎠

= 64

1
1
∴ Average number of 6 heads with six coins in 6400 throws = np = 6400 × 64 = 100 ∴ Mean of the
Poisson distribution = λ = 100 By Poisson distribution, the approximate probability of getting six
heads xtimes is given by
( )
P [ X x ] xe x ! 100

x e x !, x 0,1,2,... Problem than five b) Suppose probability (i) (ii) Solution: P [ X = To find
100

A l b) (i) (ii) Here The The ⎛ = = λ - λ = 31 a) A die is cast until 6 times? that a
 trainee soldier that the target is shot What is the probability that What is the probability that
1
Probability of getting six ∴ p = 16 & q = 1 - 6 Let x= Number of throws x ] = q x - 1 p , x =

l
1,2,3.... Since 6 can be got either in P [ X > 5 ] = 1 - P [ X ≤ 5 ] P p [ = X 0.8, = r ] q

J N
= = q 1 r5 11 - p = 0.2 - 1 pr , = 0,1,2... probability that the target would

probability that it takes him = P [ = -

appears. What is the probability shoots a target in an independent on any one shot is 0.8. the target
 =1
T
would be hit it takes him less than 5 6 first, for getting second......throws.

U
the number = │ ⎝ ⎞ │ ⎠ = be hit on the 6th attempt less than 5 shots X < ]5 on

W
shots? 6thattempt?
 o
that fashion. it must If cast the

rld
more

6. By geometric distribution = 1 - ∑
x
=

⎛│⎝x

= P [ X = ]6 = ( 0.2 ) 5
( 0.8 ) = 0.00026 5 - 6 ⎞ │ ⎠ . 1 6 = 1 - ⌈ │ │ ⌊ ⎛ │ ⎝ 1 6 ⎞ │ ⎠ + ⎛
│ ⎝ 5 6 ⎞⎛ ││ ⎠⎝ 1 6 ⎞ │ ⎠ + ⎛ │ ⎝ 5 6 ⎞ │ ⎠ 2 ⎛ │ ⎝ 1 6 ⎞ │ ⎠ + ⎛ │ ⎝ 5 6 3 1 6

5 41 6 6
⎞│⎠⎛│⎝⎞│⎠+⎛│⎝⎞│⎠⎛│⎝⎞│⎠⌉││⌋

=1-

1 6 ⌈ │ │ ⌊ 1 - ⎛ │ ⎝ 5 6 ⎞ │ ⎠ 5⌉ │ │ ⌋ 1 - 5 6

5 56
0.4019

=∑ q
4 r

p = 0.8 ∑ 0.2 r -
4 1r=1r=1 = 0.81 [ + 0.2 + 0.04 + 0.008 ] = 0.9984 Problem 32 a) If X 1 , X 2 are two

independent random variables each flowing negative binomial distribution with parameters ( r 1, p )
and( r 2, 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 10 th throw is his 5th hit, if the
probability of hitting the target at any trial is 0.5?
Solution: a) Let X 1be a negative binomial variate with( r 1, p ) and X 2be another negative binomial
variate wit( r 2, p ) and let them be independent.
Then 1

()( )
1

()
M M M
X X2 Then X1+X2 variable with r 1 r+ 2 as parameter. b) Since the 10th throw should (i.e.) the 5th

hit, the first Hence + A Problem hours.probability will Solution: a) b) Statement:

l t q pe
(i) (ii) operate A = -
t-r (t)=( q-
pe

)
-r2

l
( t ) =M X ( t ) M X ( =( q - pe t )
1 2 -
(r
1+
r
2 we component The By 33 have that N.B.D,

J N
component a) ∴ Required x it State = will 5, has r P and fail = [ an X 5,
 x1
T r
has probability = p 9 x = throws ] result q = This = ⎛ │ ⎝ 2 in should proves = x

t
U
the - P [ ⎞ │ ⎠ 5X th At 15,000 hours the for another 5000 If X is exponentially )

W
have the , ) success which result. resulted is the m.g.f in 4 = ]5 of a successes =

⎛ 5+ 5-1⎞ ⎛ 1 ⎞
│⎝ 5 │ ⎠ p 5q 5= 9 C 4 │ ⎝ 2 │ ⎠ 10
= 0.123 .
 prove the memoryless property of exponential exponential time to failure distribution with already

been in operation for its mean by 15,000 hours? component is still in operation. What is the hours.

or
distributed with parametersλ , negative binomial and 5 failures.

distribution. mean of 10,000

life. What is the

probability that it
then for any two

ld
1
pq r x positive integers s and t, P [ x > s + t / x > s ] = P [ x > t ]
Proof:

The p.d.f of X is f ( x ) = ⎧ ⎨⎩
, 0
0 λ e -λx x ≥

,Otherwise ∴ P [ X > t ] = ∞ ∫
λ
e - λ x dx = ⌈ ⌊ -e - λ x

ll
/
⌉ ⌋ ∞ t = e λ- t t P [ X s t x s ]
P[ x
P
s
[x A Problem approximately 34

J N t x s]
The distributed Daily ∞ consumption as ∴ > + > = > + ⋂ > >

P X
s]= P[ [X>>s+s]
t
]
e
= -λ

( s+

t) e- λ
s =e-
λ
t

= P [ x > t ] b) Let X denote the time to failure of the component then X has exponential with Mean
= 1000 hours.
 TU
∴ λ1 = 10,000 ⇒λ = 10,000 1 - The p.d.f. of X is f ( x ) =

Wo
distribution

rld
=∫

= - = of milk in a city in excess of 20,000 gallons is a Gamma variate with parameters α = 2 and ⎧ │
 1
⎨│⎩ 10,000x
e 10,000 , x ≥ 0 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 ] = P [ 10,000 P [ X < X

< 15,000 >

10,000
]

]
15,000

∫
( ) - 1 1.5 = 10,000
∞ ∫
()
1
10,000

f f x dx = e x dx
-
e
-

e -= 0.3679 0.3679
- 0.2231

= 0.3936 .
(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]
( ) 0.5
5000

f x dx e
0.6065
λ= . The city has a daily stock of 30,000 gallons. What is the probability
1 10,000 that the
stock is insufficient on a particular day?
Solution: 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.
1
f(y)= ( 10,000 ) 2 Γ
(2

)
y 2 - 1 e - 10,000
y

,y≥0f( y)
= ye -
10,000 y

(10,000)
,
2 y
0.
≥ ∴ The daily stock of the city is 30,000 gallons; the required insufficient on a particular day is

All
given by PX [ > 30,000 ] = PY [ >10,000 ] ()
1

JNTU ⎛ ⎞
( = μ = 0.1 - 0.5 Γ │ ⎝ 1 + 0.5 1 │ ⎠
 W o r
) probability that the stock is

ld
= ( 0.1 ) -2
Γ ( 3 ) = 100 × 2 ∞ ∞ = ∫ g y dy = ∫
10,000 10,000 10,000 ye - y
10,0002

dy
Put Z = 10,000

y , then dz =

10,000 dy ∴ P [ X > 30,000 ]

=∞ ∫
ze - z dz
1

= ⌈ ⌊ - ze - z - e
-z ⌉⌋1

∞ = 2e Problem 35 a) suppose that the lifetime of a certain kind of an emergency backup battery (in
hours) is a r.v. X , having the Weibull distribution with parameter α = 0.1 and β = 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 α = 25 and β = 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: a) For the p.d.f of X in the form f ( x ) = αβ x β - 1 e - α x β , x

> 0 Mean

= α -β1Γ ⎛ │ ⎝ β

1 + 1 ⎞ │ ⎠ i) Mean life time ( )

200hours = ii) Probability that a battery will last more than 300 hours = P [ X >

300 ] ∞ ∫ f ( x ) dx = ∞

∫
αβ
x β-1

e - α x dxβ

300 300

l
( )( ) 0.5 ( 0.1 ) 0.5
300
Al
JNTU W o
=≥ ∞ = ∫

∞ = ∫ - - Put y = ( 0.1 ) x 0.5 Then dy =

( 0.1 )( 0.5 ) x - 0.5 dx [ ]

rld
(0.1)(300)

0.5 ∫
0.1 0.5 x e x dx ∞ ∴ P X > 300 = e - y dy

= e- ( 0.1 )( 300 ) = 0.177 b) Let X be life length of the tube Then the p.d.f of X is given by f ( x ) = αβ
0.5

x β-1e -αx , x
β

> 0 i.e., f ( x )

= 50 xe - 25 x , x > 0 P [a tube is not replaced during the first 2 months]

2

⌈ ⌉ =
= P │ ⌊ X> 16 │ ⌋ , since two months
1 yrs
6 ∞ =∫50 xe - 25
x
2 dx

16 = ( -e ) 25 x 2

∞1 =

e
25
- 36

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

= ( e - 25/36 )6 = e - 25/6 = 0.0155 .

Problem 36 Support that the service life (in hours) of a semi conductor is a random variable having
the Weibull distribution with α = 0.025 and β = 0.5 , what is the probability that such a semi conductor
will be in working condition after 4000 hours?
Solution: Let X be the service life of the conductor. Then X has Weibull distribution with parameterα
and β and whose density is given by
f(x)=

αβ x β - 1 e - ax Here α =0.025 and β =0.5

β

P [Conductor will be in working after 4000 hours] P [ x 4000 ] f ( x )

dx
4000

We know that P [ X ≥ a ] =

e α- a ∴ Required probability =e- ( 0.025 )( 4000 )0.5 = e- 1.58 = 0.2057 Problem 37 a) The amount of time
β

that a camera will run without having to be reset is a random variable having exponential distribution
with θ = 50 days. Find the probability 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.
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
θ = 50 days. The p.d.f of X is given by
()
A ll
Problem y = cos xπ 38 - f x =

JNTU

World
= ∫ f x dx = 30 1 30 - 20 = 1 3 If X is a random variable uniformly distribution in (-1,1), find the
p.d.f. of Solution: ⎧ │ ⎨│ ⎩
 θ 1 e xθ

,x≥0
0x

< 0 (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 ] 20 2 50 5

0 =∫

50 1 e - x
dx = 1 - e - = 0.3297 (ii) P[Camera will not have to be reset in at least 60 days] =

P [ camera will run for atleast 60 days ] = [ > ]

= ∞ ∫ - = - = 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
()
p.d.f.

P X 60 1 60 50
x

6 e 50 dx e
5 0.3012 1 f x

= ⎧ │ ⎨│⎩

0 30,0 ,

< x < 30 Otherwise

The probability that he has to wait at least 20 minutes = P [ X ≥

20 ] 30 ()()
20

1 The p.d.f of X is f X ( x
)
x
= ⎧ │ ⎨│⎩ o 2 , , 1 1 y = cos - < < otherwise xπ dy dx = - π sin π x = - π 1 - y 2 ∴ dxdy = π 11 -
y2
Range ofY : - 1 < x < 1 ⇒- 1 < π

1
cos - 1 y < 1 ⇒- π < cos - 1 y < π ⇒- 1 < y < 1 ∴ p.d.f ofY is f ( y ) = A Find follows

l l J
Solution: b) the If the Weibull p.d.f r.v. of X (i) distribution follows Y <

NT
Y = 8 X 3 (ii) Y = 3 X + 1 . an exponential distribution with parameters ⎧ │

⎨│⎩ 0 2 π
U World
= ⎧ ⎨⎩ 1 1 - y

, - 1 < y < 1 , Otherwise

Problem 39 If X is uniformly distributed in ⎛ │ ⎝ -π 2 , π 2

⎞ │ ⎠ , find the p.d.f of Y tan X . Solution: π a) The p.d.f of X is ( )

=<.
⎧
with parameter 2, prove that Y = X 3 2 and 1/3. │ ⎨│⎩ π- < < π
y = tan x ⇒x =
1
tan - 1 ( y ) 1
2 f Xx
x
1,2 20,
Otherwise
∴ dxdy = + y ∴ p.d.f. of f Y ( y ) =
1 1
f X ( x ) dx dy = π 1+ y 2 , -∞ < y < ∞ (This distribution of y is called Cauchy distribution)
2,0 xx1
Problem 40 a) Given the r.v. X with density function ()
X
0,fx

otherwise

2 y = 8 x 3 ⇒dy dx

= 8.3 x 2 = 24 x 2 = 24 ⎛ │ ⎝ 8
y 3 2 6 y 3 ⎞ │ ⎠ = 123 6
a) (i)
 dx
β = dy dx = y - The p.d.f of Y f Y ( y ) = f X ( x ) dy

= 2 x - 1 6 y - 23 =
1 f
y 1 3 . 6 y- 2 3 ⇒ (y)
Y

y y
=16 - 13 ,0<
1
< 8 (ii) y = 3 x + 1 ⇒dy dx = 3
and x =
y -1
3

ll J
p.d.f ofY f Y ( y ) = f X ( x ) dx dy = A ∴ The
2
= αβ r.v. Y 3

NTU
( y-1) ⎛│⎝ β--α . Since x > 0 , y >
 1
World
0 = X 3 follows Weibull distribution 3

1 .⎞ 2
with parameters 2 and 3 │⎠=9 ( y - 1 ) ,1 < y < 4 b) The p.d.f of X is ( )

= ⎧ ⎨⎩ - > Then
e
f Xx 2 2x

x
, 00,
1
otherwise dy dx = 3 x 2 ⇒dy dx = 3 x

2=
1
3

y -23
The p.d.f of Y is f Y ( y ) =f X ( x ) dx dy =

13

-23 .2 - 2 1 3 y e x = - 2 3 2 1 - 2 3 ( )
y e y∴ = 2 ⎛ │ ⎝ 1 3
f y
Y ⎞ │ ⎠

y e
13-1

β = 1 the p.d.f of y is in the form f

- 2 y 13 Taking α = 2 and 3 Y( y ) y 1e
y
 l
A A

l
distribution. The two listing 1. 2. 3. mutually binomial Binomial Poisson Normal of the

J
In distributionexclusive Distribution Distribution this probabilities Distribution chapter, outcomes
 N 1 is we a for discrete discuss every of a trials. probability
T
three

U
possible These very value important distribution outcomes of a random distributions

Wo
are which appropriately variable is is called a probabilit used when there are

exactly labeled and

rld
. This is used to obtain the probability of observing x successes in n trials, with the probability of success on a single tr ial
denoted by p. The binomial distribution assumes that p is fixed for all trials. Consider a random experiment which has two
possible outcomes
. The probability of success in each trial is p and remains same for all n trials. The trials are independent. Let X is a random
variable which represents the number of successes in n trial. Then X has n values named as 0 success, 1 success, 2
successes, ..., n successes., i.e.
X x : 0,1,2,3,...,n
1This distribution is associated with the name of James Bernoulli (1605 1705) and was published in 1713. It is also called as
Bernoulli distribution in his honor.
and
with probability of x successes as
P X x nC x n x
xp q

where q 1 A p: the probability of failure in a single trial.

If the random experiment of n independent trials is repeated N time, then the expected number of trials showing a success is
N.nCxpxqn
x
The set of the numbers of success x along with the corresponding probability is nown as i nomial Probabilit istribution.
The set of the number of successes along with the corresponding theoretical frequencies is called heoretical inomial
istribution with parameters n and p.
 ll
Binomial 1. 2. 3. 4. 5. The The The There successs In The a distribution

J
single Bernoulli number trials probability are and are two trial n repeated failures.

N
trials is mutually probability of of based success trial are under on independent. is
 T
exclusive finite the (and of identical success following and therefore possible fixed.

U
is conditions. p assumptions for and outcomes failure) the probability remains of each same of trial

W
failure for which each is q are where trial.
 o
referred p q to as 1.

rld
E AMPLE 11.1
If on an average one ship in every ten is wrecked, find the probability of arrival of at least 4 ships safely out of 5 ship
expected to arrive.
Solution: Given that one ship out of 10 is wrec ed. i.e., 9 are safe.
9
Probability of safe ship arriving: 10 Here, we have to compute probability of arriving at least 4 ships safely out of 5
expected to arrive. Here n is 5. Since we need at least 4, it means there are either 4 safe ship or more than that. Thus here x
has two values 4 and 5.
P 4or5 P 4 P 5
5 C p4q5 4 5C p5q5
4 5
E AMPLE 11.2
10 of screws produced in a certain factory turn out to be defective. Find the probability that in a sample of 10 screws
chosen at random, exectly two will be defective.
Solution: Here
p 10 0.1, q 1 p 0.9, n 10, x 2
5
5
910
4

110
910
5
7
5

910
4

A E AMPLE 11.3

Find out the Binomial Distribution to be expected by tossing 4 coins 320 times.
Solution: Let
X Number of successes (i.e. The number of heads when 4 coins are tossed), n 4, Number of coins, p Probability of
.
getting a head on a coin Possible q 1 values p of 12

1 and,
ll
X are : 0,1,2,3,4. The corresponding 2 probabilities are obtained of p q n:

JNT U 4
p q p4 4C p3q 4
1 C2p2q2 4C3pq3
 W 4 in the 4
q

orld
expansion

1 4 6 4 1
16 16 16 16 16 Thus, we have
X x 0 1 2 3 4 P X x 16 116 416 616 416 1Expected Frequency NP X x ,where N 320. The required theoretical
Binomial Distribution is as follows:
X x 0 1 2 3 4 Expected frequency 20 80 120 80 20

Let the Binomial distribution be

p x P X x nC x n
xp q
1
12 2
4

12
4
4
12

P 2 nC 2 n2
2p q
1 0C 0.1937
2 0.1 2 0.9 8
3
12
6
12

x, x 0,1,2,...,n
3

12
2
4
12

12
3

12

ll J
A 1 np q p 1 Variance,

N
Standard Deviation
 T U
Variance npq p npq

World
Let the Binomial Distribution be
P X x nCxpxqn

x, x 0,1,2,...,n
Mean,
Mean, X X

2 X

n x 0xp x

nx 0x
nC x n x 0.qn 1.nC1qn
xp q

2p2 ... npn

0 nqn

1p2 ... npn

np qn
1 np q p n
X 2 X2
X 2

Now
X2

nx 0x
2p x

nx 0x
2 nC x n
xp q

2 X2
x

n x x
x

x 0

nx 0x x
x

nx 2n n

x np
n n
1 np
n n
X 2

n n
np 2 n2p2
n2p2
np
1 n
1p n n
X np
np2 np 1
1 p2 q p n
1 p2 np q p 1
1 p2 np
1 qn
np2 np

1p 2.nC2qn
q
1 1! n

1 .nCxpxqn
1 p2.n

1 x .nCxpxqn

2p ... pn
2Cx

px
2

n
n x 0x. Cxpxqn

2qn

Standard deviation,
npq
First central moment, E AMPLE 0 Second central moment,
2 npq
Third central moment,
p
Fourth central moment,
6pq
Karl Pearson coefficients,
The sign of

l
Al Case Case The Solution: The deviation 12

value If If mean np np of
 J
Given: x and of integer,then for p 11.4 a mode which integer, binomial npq 6pq p of

N 3 i,e.,pq
P X the mode then 0,positive distribution distribution. x there np is maximum

T ,the distribution
i,e. are 16 mean is Mean p q two 4 then and modes is called third np s ewness
U is lepto urtic.
np npq central 4 mode. 6pq p is and positive. moment np

W o rld
p is 0.48. Find standard (11.1) 1 Case 3 .If

np p k d where k is an integer and d is a fractional,then mode k.

If
If
If
If
If
If
1. where p np p p np p
3 1
2

2 3 1

2 3 1
3 npq q
3 npq q
npq q
1 depends on
3i,e.,1
6pq npq

3i,e.,pq 16,the distribution is meso urtic.

6pq npq

p 0,negative i.e, p q then no s ewness is negative.

p 0,i.e,p q then no s ewness the distribution is symmetrical.
3. If
1

3 0 then
4 3n2p2q2 npq 1

6pq 0 i,e.,pq 16,the distribution is platy urtic.

2

23

32
2
q

npq q
3

2
2

24
4
1
,
p 2 npq
1
3
1
3

6pq npq
1. If
3 0 then
1
1
p 0.48 (11.2)
Dividing equation (11.2) by (11.1), we get
npq q

A 50q2 10q 5q
 ll JN
5q 3 03 03 0

4 4
Now and n Standard q

Deviation 52 10

T
npq

U 2
35 5
World
2 4 2
25 5 5 Hence Mode 4 Integral part of np p

1. Binomial distribution is a discrete probability distribution based on Binomial Theorem.

2. Theoretical Binomial distribution can be obtained by multiplying the probabilities by N.
3. The shape and location of binomial distribution changes as p the probability of success
changes for a given n. When p q 1
p the distribution is symmetrical. When the number of success are in ascending order, the distribution is negatively s ewed
for p 0.5 (or p q) and the distribution is positively s ewed for p 0.5(or p q). That is binomial distribution is asymmetrical
for p
q. This distribution is platy urtic for pq 16, meso urtic for pq 16 and lepo urtic for pq 16.

3
q 25 50q2
3 1 5q
p
3
25
q q
30q 5q
3
q 25
q q q
3 10q 1 0
5q
1
25q
3
1 25
2q2
0.48 48
p np 4
12
400
3 q q
100 25
3 since q can not be negative
110 q 5 Hencep 1
25 2.4 1.55
np p 10
3 npq q
 3 or q
3 0 or 10q 1 0q 5
q 1
10
35
4. If n is large and if neither p nor q is too close zero i,e., the difference between p and q is very small, then the binomial
distribution tends to symmetry and may be approximately normal distribution 2. 5. The Mean constants np, Standard of

binomial deviation distribution npq, P X First central x n Cmoment,

xpxqn x,x 0,1,2,... are as follows: A npq 6pq

.

l
The use of binomial distribution is made in case of division by dichotomy where one event is called success of

population The it tool 11.1 chances disease? 11.2 ment 11.3 are 11.4 five 11.5 and 11.6 1024 tors 11.7

l
etarian whether are 11.8 is 11.9 11.10 has the binomial 0.4096 vegetarian? calculate or would binomial in with
investigators probability been six? forecasting is The The In Six Eight Assuming Assuming In Find and of they Four 12 a an 6
and
and you suffering with very dice normal binomial incidence another distribution animals all army are its coins distribution

J assuming
dice 0.2048 expect the are widely replacement means vegetarian. of that that of each battalion

were are expected rate thrown being event from caught the distribution of to respectively. half 50 tossed and applied take

N that
of is thrown events report occupational is it. 0,1,2,3,4,5 is infection nown. the standard 729 very

out
infection. 3/5 of How called 10 frequencies What or and population a that based individuals is times. sampling of 64
of 100
useful population consisting many quality this is failure. It the Find three times. of deviation. the is on Find

T
experiment disease a soldiers married How in applicable investigators certain the of random probability or
 investigators
control. decision from is Under Spot the to a less of parameter many in vegetarian theoretical find in
probability 5 a are soldiers an disease people 3 independent town sampling. an certain is Binomial out or times to infinite

U each
ma ing married industry that repeated would 4 the if reads p so was in read in binomial they

ta es
do out of situations assumptions animal a that of population situations and expect this you row considered
distribution newspapers. of newspapers the is read trials, 100 chance six the such binomial expect of observed is distribution
a sample
times. newspapers,how to wor man 5 remainder of nown probabilities in that soldiers? report with random and

W
of business. at a is Find an and the distribution. success least result. circumstances considered to or

of 10 individuals
that individual 4 wor man be further 2/5 128 without binomial sampling or three three 40 . of One more

and unmarried. 12 many 1 dice assuming and or area spot replacement. an In being 12have will distribution less from an
the 5 important
 o
investiga to 2 in 1 success contact

experi
to
show people a nature a which or What finite
20
veg that
see
2aa

rld
failure. The outcome of a 5 or 6, was not considered a trial. The results were:
2Discussed in next sections.
npq ariance, Third central moment,
2

4 3n2p2q2 npq 1
6pq ,
p 1
q npq n

3 1
2

npq q
3

p , Fourth central moment,

, 0, Second central moment,
Number of success 0 1 2 3 4
Frequency 0 5 13 22 24
11.11 Find the mean and standard deviation of the observed frequency distribution and the theo retical binomial
distribution.
2
11.12 For Theoretical Binomial distribution x np 4 12 11.13 Assuming head as a success in 512 tosses of 8 coins
together find the expected frequencies obtaining the probabilities of various successes.
No. of Success x Probability p(x) Expected Frequency Np(x)
0 1/256 2 1 8/256 16 2 28/256 56 3 56/256 112 4 70/256 140 5 56/256 112 6 28/256 56 7 8/256 16 8 1/256 2 Total 1 512
11.14 A set of 8 coins is thrown 256 times and the number of heads appearing in each throw is recorded. The results obtained
are given in the following table. Estimate the mean number of heads and then calculate the expected frequencies using this
estimate if the binomial law holds :

l
A 11.15 The 11.16 follows:

l
Estimate 11.17 11.18 11.19 probability b) a) c) There In Obtain In 4 the n 99, n 60, n 8 coins a 104
 J
probability binomial are , of the p 1/5 liters are p 0.6 p 1/6 No. a 64 No.of mean tossed flower beds

N
distributionmeanis of of of No. 4 throws heads and 160 in a being mice, No. mouse of a mode garden

T
times female of blue the 0 2 litters,f being for and number is and mice,x the 1 6 1/4. following a 3 3 following

female and 38 seeds Find 2 of variance 0 8 litters the 52 of 3 and particular binomial number 28 1 find which 59 is 4 2, the
34 2 find of 56 5 distribution contain type expected beds the 24 3 32 of 6 remaining with flower 0,1,2,3,4 10
4
frequencies. 10 7 3,2,1 if 8 1
are sown in each bed. and 0 blue flowers.
female mice is as

U W
constants ( observed frequency distribution:

orld
).
2

No. of heads 0 1 2 3 4 No. of tosses 17 32 54 51 6

,
1

,
2

,
3

,
4

,
1

Assuming that the coins are unbiased calculate the expected frequencies for the number of heads and test the goodness of
fit.
11.20 A set of 5 coins was tossed 96 times. The following table gives the observed frequency distribution of the number of
heads in a single trial. Calculate the expected frequencies on the assumption that the coins were unbiased. use test to
examine the correctness of this assumption (x number of coins that show head in a single trial, f No. of throws)
x 5 4 3 2 1 0 f 7 19 35 24 8 3
11.21 The observed frequency distribution of 128 throws of 7 coins, according to the number of heads. Fit a binomial
distribution under the assumption that the coins are unbiased. What are the mean and standard deviation of the fitted
distribution?
No. of heads 0 1 2 3 4 5 6 7 No of throws 7 6 19 35 30 23 7 1
11.22 Criticize the statement: The mean of a binomial distribution 8 and standard deviation is 3.

Let X be a discrete random variable where probability of X x is given by

e
P X x x! mmx

,x 0,1,2......., m
A the probability distribution of the distribution. Poisson small and n is large enough

ll J
Theorem 13. n binomial i.e., n , in such a wa Proof. Probability of x successes

N
that distribution of such distribution P X X np is that in said x binomial m np to p q is
 T n some be x! n Ca m n!
P X is xqPoisson limiting (Constant). distribution nn , fixed x when

U e
x distributionform of binomial 3 p is ver small, number, then mmx x!

is

px
xpx

x! n
n!

Wo
0
with parameter m. Here m is mean distribution where p is very
i.e., p 0 and n is ver large, the probability of x successes is

rld
m
x!
xx

when n
m e
x! x
m

3This distribution was developed by S.D. Poisson in 1837.

1
1
x !
x !
mn
1
n n!

n
p n

mn
x !
n
x
1
mn
mn
as when n tends to infinity, we have
n lim 1
m
When the expected frequencies for an observed frequency distribution are obtained using Poisson
 Law, then general expected frequency is given by N.e

m
A
, where N
The theoretical distribution so obtained is called

total observed frequencies.

theoretical m x! xPoisson
distribution. Poissondistribution is applicable in caseof a number of random situations with a small probability of success.
Poisson distribution is used in place of binomial distribution when p is very very large and np m is fixed number. Some
very common cases where Poisson distribution is used: The number of deaths at any year by an epidemic or by a rare
disease;
1. The number of mista es per page by an experienced typist;
2. The number of typographical errors per page in a boo printed in a good press;
3. The number of defective items in the items manufactured 4. The number of bacteria in a drop of clean water;
5. The number of suicides per day in a city;

ll
6. The number of telephone calls per minute in an telephone Mean:

J N T x.p x
X x 0 x 0
p x
U
x 0 by booth.

W
a big factory;

o
small,n a place in is
rld
xp x
x 0x.e

m.m.em m
Second Moment about origin :
2

x2p x x 0

mx
x x x 0m x!

x 0x x

mx
m x!
m2e
x
2 !
m2e

m.em m m2 m
mx
m x!
e
1 !
e

m x 2

m.m x 1

1 e
1 x e
mx
2
x
mn
 mx
m x!
1
x
mx

n e

x 0xe

ll JNTU
A
World
Second Moment about mean:
2

m2 m
Variance:
m
2 Standard Deviation:

ariance m

Mode: case 1: If m is a positive integer, then there are two modes at x x

1 and x m. case 2: If m is not a positive integer, then the integral part of m is mode.
Third moment about mean,
 1
3 m Remark: The advantage of Poisson distribution is that if we now the value of the parameter m, we can find all the
other constants of this distribution.
E AMPLE 11.5
The parameter m of a Poisson distribution is 3.24. Find the constants of this distribution.
Solution:
m ,x 0,1,2,.........
P X x p x em x! x
p x e
3.24
3.24 x! x
,x 0,1,2,.......

Mean , X m 3.24
Variance,
2 m 3.24
Second central moment,
2 Standard Deviation ,

m 3.24 1.8
Third central moment,
m, always positive
3 Fourth moment about mean,
3m 1 3.24 3 3.24 1 3.24 9.72 1 3.24
4

1
3m2
4 m or m 3m 1
22
4
2

23

32
m 3m 1

3m 1
m2
3 1
m m
m2 1 ,
m3 m
2
2 m2 m
1

1
3 3 m

1
1
1 m
10.72 34.7328
A Mode:

Since m is not a positive integer, E AMPLE 11.6

In a certain Poisson frequency the frequency corresponding Solution: Let the total frequency tion is
Frequency of 2 successes N.e

ll
Frequency of 3 successes N.e According to the condition given:

JN mm
to hence m m N.e distribution Ne be 3 2! 3! successes. 23N. mmode m
 T m , x
.Then x! 3 x3.24 the the 3

1 0,1,2,3..........
U 1
m Find frequency frequency 3 0.24 its 3.24 mean corresponding for

.3086

W
and x standard successes deviation.
 o
to in 2 the successes Poisson is distribu half

rld
6
1

Mean, X m 6 and Standard deviation

m 6 2.449.

In a Poisson distribution the probability of x successes ( or probability mass function of X) is given by

p x e
m for the various values of m are given in a table from which they can be read easily or the values of m are given in a table
from which they can be read easily or the values may be computed as follows:
e
xx! ,x 0,1,2,...
m .m To find the probabilities for x 0,1,2,... first of all we obtain the value of em.The values of e

2 Antilog
1 3 1 3 .3086 3.3086
2 3 m 3.24
1

2
2log10e Antilog
1m
22! 1 Ne
m. m 2
1 1 .3086
1 m 3.24
m3
m 3! 12
2
1
1 2
2
1m

13.24 0.3086 0.5556

132
m3
2
2
2log102.718 0.1353
After finding the value of e m or p(0) we find p(1), p(2), ... etc. By the relation
p x e
xor by the relation p x 1 m
m.mx! x 1
p x
Expected frequencies are obtained as follows: Expected frequency of x successes Np x .
m0123456789
0.0 1.0000 .9900 .9802 .9704 .9608 .9512 .9418 .9324 .9231 .9139 0.1 .9048 .8958 .8869 .8681 .8694 .8607 .8521 .8437 .8353 .8207 0.2
.8187 .8106 .8025 .7945 .7866 .7788 .7711 .7634 .7558 .7483 0.3 .7408 .7334 .7261 .7189 .7118 .7047 .6977 .6907 .6839 .6771 0.4 .6703
.6636 .6570 .6505 .6440 .6376 .6313 .6250 .6188 .6126 0.5 .6065 .6005 .5945 .5886 .5827 .5770 .5712 .5655 .5599 .5543 0.6 .5488 .5434
.5379 .5326 .5273 .5220 .5169 .4117 .5066 .5016 0.7 .4966 .4916 .4868 .4819 .1771 .4724 .4677 .4630 .4548 .4538 0.8 .4493 .4449 .4404
.4360 .4317 .4274 .4232 .4190 .4148 .4107 0.9 .4066 .4025 .3985 .3942 .3906 .3867 .3829 .3791 .3753 .3716
The value of e

ll J
To read then read the Similarly e A the value value in m of column e
N T
0.002479 m for (1) m in front 0.000912 0.51 of e from 0.5. 0.51 the Thus

U W
0.000335 .6005 above table, 0.000123 we ta e 0.000045

o
0.5 in the column m and
rld
E AMPLE 11.7
Suppose that a manufactured product has 2 defects per unit if product is inspected. Using Poisson distribution calculate the
probabilities of finding a product (i) without any defect , (ii) 3 defects (iii) with 4 defects , (iv) with 3 defects or 4 defects . (Given
e 2 0.135).
Solution: Let X Number of defects per unit, and
p x P X x e
m ,x 0,1,...
m x! x
Here m 2
p 0 e
0.61 .5434 and e
mfor m 1 to m 10
m12345
e

m 0.36788 0.13534 0.04979 0.01832 0.006738 m 6 7 8 9 10 e

2 0.135 given
p 1 p 0
2 0.27
p 2 p 1
22 0.27
p 3 p 2
2.51 e
2.0

0.51 .13534
m 0.135
m2 0.27

m3 0.27

23 0.18
.6005 .08127
24 0.09 In this way (i) The probability of no defect p(0) 0.135 (ii) The probability of 3 defects p(3) 0.18 (iii) The
probability of 4 defects

11.8
p(4) 0.09 (iv) The probability of 3 or 4 defects 0.18 0.09 0.27 A E AMPLE

2 percent of the electric bulbs produced by a company are defective.Find the probability that in a sample of 200 bulbs (i)less
than bulbs are defective and (ii) more than 3 bulbs are defective. (Given e 4 0.0.183)
Solution:
2 .02
n 200, p 2 100
m np 200

l
(i) P (less than 2 bulbs (ii) P (more than 3 bulbs P X E AMPLE 11.9 Solution: number The mean (i) accidents

l J
who distribution of of taxi are X the in driver not a Number distribution year. X x involved of 0
 N
are accidents 3 are P X defective) defective) is in 1 3. any by 2 Find accident taxi P X

T 4 ,x
U
out drivers the in 0 P X 4 x! a member in xyear city 0.0183 and

0,1,2,..........
W
in of 1 a (ii) taxi year 0.0915 who e drivers is a are Poisson involved out of of

,x 0,1,2,3,......
accidents p x P X x in a e year m mx! m x 3, the mean of Poisson Distribution Probability,
P(X x),p(x) Expected Number , Np(x)
e
 o
Distribution. 1000 in The taxi drivers more than 3

N 1000 ,

rld
0.2241 224.1
Figure
4.4
33!

1
0.0498 49.8 1 me
0.1494 149.4 2 m2e
0.2241 224.1 3 m3e
e
4. 41! e

4 1 4 8 323 1

4 1 4 5e
m 3! 2! m
27926 e e
.02 4
P X x e
m x,x 0,1,2,........
m x!
P X x e
m e
m 3e
m 0.18
p 4 p 3 4
P X 3 1
3 0.0498 1000
3 0.1494 1000
3 0.2241 1000
3 0.2241 1000
4 5
.0183
e
4 e
713 1

4. 41! e

4 e
.4331 .5669
22! e
4.4

E AMPLE E AMPLE 11.10 A manufacturer of pins nows that on an average 5 of his product is defective.He sells pins
in boxes of 100 and guarantees that not more than 4 pins will be defective. What probability that a box will meet the
guaranteed quality?(e 5 .0067)
Solution: The probability that pin is defective , p 5 or 0.05
Let X The number of defective pins in a pac ets of 100 pins, p 0.05 and n 100,
m n

l
100 5
Al and Solution: In be the defective a defective. approximate certain blades,
J
factory 11.11

N
The number respectively blades turning of are out pac ets supplied in razor a p x consignment

T U
containing blades, in P X pac ets there x (1) of of 10000pac ets. is no
 W
10. e a defective, small Use chance Poisson (2)one of distribution 1/500 defective for any and to is

o
the blade to calculate (3) two

rld
0, (i) Np 0 49.8 50
no accident (ii) N
54 .1746
P(a box will meet the guaranteed quality)
P (The number of defective pins is not more than 4) 1
N p 0 p 1 p 2 p 3 , more than 3 accidents
1000
p (The number of defective pins is 4 or less) 1
p 0 p 1 p 2 p 3 p 4 1
0.0067 0.0335 0.0838 0.1397 0.1746 0.5617
m. m
xx!

p 0 e
49.8 149.4 224.1 224.1 1000
5 .0067
p 1 p 0
51 .0335
p 2 p 1
52 .0838
p 3 p 2
53 .1397
p 4 p 3
647.4 352.6
p 0.05
m1 .0067

m2 .0335

m3 .0838

m4 .1397
35.3
Here,
1, N 10000
n 10, p 500
1 .02
m np 10 500 Let X The number of defective blades in a pac et ( 0,1,2,3,4,5,6,7,8,9,10).
According to the Poisson distribution,
p x P X x e
0.02 .9802
p x 1 p x
gives,
mx 1
p 1 p 0
0.021 0.019604
p 2 p 1
p 3 p 2
Required expected frequencies: (1) No defective e0 Np 0 10000 0.9802 (2) One defective e1 Np 0 10000
0.019604 (3) Two defective e2 Np 2 10000
E AMPLE 11.12
 l l
A Solution: (i) (ii) tion) Neither A on days Let The car each X demand on hire car

J
which demand day firm is is is used (i) distributed p X has refused, of neither i.e., a two car,

N
2 the cars means thencar as demand 1

1
1
 T
1 p X which Poisson is p X used, the p 0 p X x is demand it 0 and hires zero,

U m ,
distribution 2 p 1 e 9802.0 196.04 (ii) out some day with demand by m x! x

where m2
required probability e 1.5 is more than p 2 m 2! 0.029802 2 196
0.00019604

W
0 day. The number of demands mean 1.5. Calculate the is refused.

m 1.5.
(or proportion)
 .2231
2: The required probability

o
for a car proportion of

(or propor

rld
0.1913
e
0.2231 0.2231
mx , x 0,1,2,........
m x! P 0 e

m em m e
0.021 0.9802

0.022 0.019604

0.023 0.00019604
0.00019604 1.9604 2
m e
1.5 0.2231
0.023
1.5 2 2
E AMPLE 11.13
Mean of a distribution which is explained by the Poisson distribution is 2.0. Find out the probabilities corresponding to various
successes 0,1,2,3........ If the total frequency is 100 what are the various frequencies?
Solution:
Probability of r successes
m , where m 2
p r em r! r
p 0 e

A From formula p r 1 p r

r 1m, we get,
p 1 p 0
p 6 p 5

l l
The are less. We is obtained If decreasing have computations Thus the p x x not total the by

J
computed frequency Poisson subtracting very . 1353 0 may fast distribution the and be . is
N
2706 the 1 100,then probabilities carried will p 7 sum become . p 0 P 1 is out 2706 the 0 x

T U
2p 6 as up follows: expected after very to . infinite 100 1804 p(7), low 3 after

successes, 2 .03608 .01203 some 0.1353 .0902 time, .0120266 .0034371 but .03608 then it is

W o
the clear computation that the probabilities is meaning 4 5 6 7 or more . 0902 . 0361

. 0120 . 0034
the last probability is for 7 or more than 7s. this p 2 p 3 p 4 p 5 p 6 from 1. frequencies are as follows
:
Np x
rld
.0120 1.20 1 7 or more 100
.1353 .2706
p 2 p 1
.2706 .2706
p 3 p 2
.2706 .1804
p 4 p 3
.1804 .0902
p 5 p 4
.1353 13.53 14 1 100
.2706 27.06 27 2 100
.2706 27.06 27 3 100
.1804 18.04 18 4 100
.0902 9.02 9 5 100
.0361 3.61 4 6 100
m1 2p 0 2
2
2.71828
2
m6 6
2
m7 7
m0 e
m 0!
2
m5 5
2
m2 2
2
m3 3
2
m4 4
.0034 0.34 0
m e
E AMPLE 11.14
The following table give the data regarding the deaths of soldiers from ic s of horses per army corps per year for General
Army Corps:
Deaths 0 1 2 3 4 Frequency 109 65 22 3 1
Fit a Poisson distribution and calculates expected frequencies.
Solution:
Death x Frequency f fx
0 109 0 1 65 65 2 22 44 3 3 9 4 1 4
Total 200 122
Mean ,
fx
m
122
f
0.61,e m
200 e

ll
Formula A of 4. To p make (4 or :

J N
the total probability p 1 p 3 p 4 p 1 P X 1, p 2 p 2
 T m
U ,x
p 3 p 0 p x 1 we can x find ep x the m x! probability x

0,1,2,....
x 1m
,gives
0.61 0.610.610.612 3 4 0.61 of ‘4 0.3314 0.1011
0.0205
0.0031

0.5434
and more)
1
p 0 p 1 p 2 p 3
1

World
more’ instead of probability
.9964 .0036
Remark: The number of deaths of German soldiers in 20 units of German Army Corps within 20 years from ic s of horses
is an important example of Poisson distribution. Probabilities and Expected frequencies are as follows:
.5434 .3314 .1011 .0205
1
0.612 0.3314

0.613 0.1011

0.614 0.0205
0.611 0.5434

A 0.0031 0.6 1 Total 0.9995 1 200

l
11.23 than 11.24 the e 11.25 follow 11.26 bution 11.27 there errors 11.28 and fail 11.29 basis 11.30

l
Poisson telephone 11.31 the will 11.32 be approximate 4 defective. value probability to guarantees have

two are of in meet .018) find Poisson b) a) c) Poisson distribution If The In The After A A The The In the 3 defective
probability a call 2 up more errors telephone the producer a the reaction the a whole or town number past average
 J
probability number The to certain and correcting probability proportion less distribution that guarantee? of 4

than distribution per experience calls 10 lenses (ii) the in boo calls not from that claims with 5 of a accidents 4 exchange

N
factory number of exactly occurrence pages sample calls per more pac ets per out of that telephone are

injection. the m of that minute, , 1,000 that per of find of minute, shows . supplied defective a turning than errors of 3 Use
of during 1000 ta e particular receiving: receives 3 calls containing minute. 5 customers the 10.(It . pages of 3 Poisson

T
and that place What call of of items persons probability in less out one in items his the is there the (Givene

on an (Given: pac ets injection in optical nown than is particular will product first distribution an (i) same per operator a (
the occur in span average no be 4 minute 50 a that probability e of time that defective, lenses industrial .6 bulk defective of
 U
will is pages on 4 of minute there receives 50 defective. e an interval .5488 4 10. is arriving to

have .0813) days.Assuming calls 4 there average 4 of estimate will Use percent (ii) accidents . exactly .6703) a that

reaction ) per What from is tomorrow? book, be one Poisson at He a minute. 4 the three a the small industrial sells is

W
defective, find 6 8 shop it operator to customers the in to number is or that an the distribution his a 5

chance Find is probability found more certain individual minute 5. product probability the accidents ) (iii) at of the (i) Using
for
number accidents that will least pages 500will probability two 1month. past in to on arrive that Poisson one is not
any
defective per boxes calculate with of of 0.002. 8 the a individual in month.Find follows accidents receive not (Given .

o lens
box average 0,1,2,3, a of on distri day. more
Find will 100
(iv) the
to
the aa:

rld
three defective and (v) four defective lenses respectively in a consignment of 20000 pac ets. (given e
.02 .9802)
x p x Np x
0 0.5434 200 0.5434 108.7 109 1 0.3314 200
0.3314 66.3 66 2 0.1011 200
0.1011 20.2 20 3 0.0205 200
0.0205 4.1 4 4 or more 0.0031 20
) exactly 2 individuals and (
11.33 In factory manufacturing fountain pen, the chance having a defective pen is 0.5 percent, 100 pens are ept in box. What
is the percentage of boxes in which (i) there is no defective pen (ii) there is at least one defective pen and (iii)there are 2 or
more than 2 defective pen?(given e .5 .6065)
11.34 A manager accepts the wor submitted by his typist only when there is no mista e in the wor . The typist has to type
on an average 20 letters per day of about 200 words each. Use Poisson distribution to find the chance of his making a mistake
if (i) less than 1% of the letters submitted by him are rejected and (ii) on 90 of days all the wor submitted by him accepted.
11.35 250 passengers have made reservations for a flight from Delhi to Mumbai. The probability that a passenger who has
reservation will not turn up is 0.016. Find the probability that at the most 3 passengers will not turn up.

l
A 11.36 data: (Given: 11.37 Poisson (given 11.38 distribution 11.39 7. The considered
 l
Alternatively tion curve. quency Karl Later used p nEven The when when perfectly Gauss. and

on e Such curve.It normal e distribution The A Show Comment French though developed .1 n neither

J
.5 as typist approaches is curves This following we .3679) the 16 smooth .6065)

that distribution was Mathematician p can and is limit commits p and are on why nor by there and extensively say the

N
the No. and also Quetlet, toward q mista es infinitely q calculate it Mista es

that standard are is is following called of symmetrical called the No. was very a No. not the mista es inconsistency
which following Galton of Pierre developed first per small. of normal Gaussian equal, Normal is theoretical
 T
deviation per pages pages statement: known page discovered the page S. per and for curve, curve binomial

Laplace Probability number were curve page Fisher. and is as very frequencies: in 9 . 42

U
represents the 0 For utilized observed resulting the by in large developed of distribution 200 normal

his following the 0 33 a 1 mista es curve Poisson n honor. by English we in from a 90 14 1 2 German curve. continues or a
get this boo . Normal distribution statement: approaches per the 20 6 3 2 Mathematician perfectly principle. Thus,

W
expansion page Mathematician Fit 4 4 10 and curve 3 a in 1 5 Poisson the infinite The mean

smooth as 4 0 typing This of normal n of Error increases Dr. mean is the principle binomial distribution 8 and 100 and and or
 o
A. binomial curve of symmetrical Normal pages Artsonomer De variance to a was distribu- Moivre. Poisson

infinity. may to .Fit q Fre

also the
be is a

rld
The following are the four fundamental assumptions or conditions prevail among the factors affecting the individual events
on which a normal distribution of observations is based:
1. ndependent actors: The forces affecting events must be independent of one another.
2. Numerous actors: The causal forces must be numerous and of equal weight (importance).
3. S mmetr : The operation of the causal forces must be such that positive deviations from the
mean are balanced as to magnitude and number by negative deviations from the mean.
4. Homogeneit : These forces must be the same over the population from which the observations
are drawn (although their incidence will vary from event to event).

We obtain the theoretical frequencies for X 0,1,2,3,.........n using the expansion N q p n.if p q then binomial distribution
becomes a symmetrical frequency distribution. When n is very large it is very difficult to compute the expected frequencies.
This difficulty is overcome by normal curve. Normal curve is a continuous algebraic curve. Its formula is
or
N
2 A 2

Ni
where computed height of an ordinate of the of x from the mean.

ll
N number of items in frequency distribution, i width of the class interval,

JNT U
2
curve at a
 W o
given point of x series i,e. at a distance

rld
Standard deviation of the distribution,

1. Bell shaped : The normal curve is a bell shaped curve and describes the probability distribution
of a normal variate with two parameters
(standard deviation).
2. Continuous : Normal curve/Normal distribution is continuous.
3. Equality of central value : Being perfectly symmetrical values mean, median and mode of normal distribution are same.
The ordinate of normal curve is maximum of the central value.
4. Unimodal : Since the maximum ordinate in normal curve is at one point only, hence normal
distribution is unimodal, and normal curve is single humped.
5. Equal distance of Quartiles from median: The difference between third quartile and median
Q3
Q1 . 6. Parameters and Constants : The two parameters of the normal distribution are mean (
3 0
Odd central moments are always zero.S ewness is zero and the curve is meso urtic.
7. Points of Inflexion : Near the mean value, the normal curve is concave while near
or X) and standard deviation (
it is convex to the horizontal axis. the points where the change in curvature occurs are
is equal to the difference between median and first quartile
). The other constants are : Moments :
1 0,
1

1 0,
2

23

32
0,

2
2

2,

(mean ) and
3 0,
e
2
2
e
x
12

3 3
2
x 2

23
4
x
4 3
3

23

22

22 3
3,
4,

1
3
.
8. Area under the curve : The area lying between the normal curve and the horizontal axis is said to be the area under the
curve and is equal to 1 in case of probabilities and is equal to total frequency in case of frequencies in the distribution.
Based on mean and standard deviation the specific area under normal curve are as follows:
(a) With in a range 0.6745 of on the both sides the middle 50 of the observations occur

) covers 68.268
i,e. mean

area, 34.134
A covers 50 area 25 on each side. (b) Mean

area lies on either side of mean. (c) Mean

 ll
) covers 95.45 area, 47.725 area lies on either side on mean. (d) (e) Mean of Only

J N T
the 0.27 mean. area is outside the ) covers range

U W o
99.73 area, 49.856 area lies on the either side
rld
It is useful to transform a normally distributed variable into such a form that a single table of areas under the normal curve
would be applicable regardless of the unity of the original distribution.
3 S. . (i,e.
1S. . (i,e.
2S. . (i,e.
0.6745
25 25

2
1
47.72 47.72

34.13 34.13

2
3

.
0.6745
3
0.6745
0.6745
2
2

A Let X be random variable distributed normally with mean define a new random variable as
X

where New random variable, X value of X, X Mean a standard normal variate and its distribution is called
standard standard deviation unity. Similarly we can obtain the values in figure 3: In the table “Area under Normal Curve” first
column is 0.0 to 3.0. In next, there are 10 columns representing the numbers given. Suppose the value of 1.45. Then
 l
read the area corresponding 0.4265. Thus the area from 0 to 1.45 is 0.4265 or 42.65 . of 0.4265 The 1.45 mean

l J
area from and and between 0.5000. for 2 1.05 negative 1 is Thus 1.45 obtained area

N
area the area area and from from area from to as required 2 the is follows: 1.45 1.45 0 to 1.05
 T
right to left 1.05 area to to of is to 0 0 1.45 obtained mean. 0.4265 0.3531 0.4265 0.3531 0.4265 is

U
0.5000 0.4265 To as find follows X and of X, normal of for in which For positive the area :

W
0.00 any standard S.D. the from 0.0735. distribution given to 1.4 numbers 0.09 1.45 of the

under deviation
The value X. area in to with Then area are which infinite 0.05 of is written X mean between to where , as areas the
o
is then subtract shown called 0 from

right it and are we

is 1
0.7796

rld
area from
E AMPLE 11.15
How would you use normal distribution to find approximately the frequency of exactly 5 successes in 100 trials, the
probability of success in each trial being p 0.1?
Solution:
Let the number of trials n, then n 100, p 0.1 and q 1 p 1
0.1 0.9. Mean in case of binomial distribution
np 100
0.1 10
and
1.05 to 0 0.3531 required area 0.4265
npq 100
33
49.85 49.85

0.1

X
0.9 9 3
0.3531 0.0734
3
When n is large, n binomial distribution tends to normal distribution which is continues. Hence, the frequency of 5 success will
be in the class 47.5 5.5 and the mean and standard deviation of binomial distribution will be the mean and standard deviation
of normal distribution.
4.5
for X 4.5, 1
Area between 1.83 and 1.5 frequency of the class 4.5
E AMPLE 11.16
The mean of a distribution is normal, what percentage beyond 72 and (iv) between Solution: (i) Between 60 and 72 :

ll JN
A is of 70 60 0.4664 items and with 80?

T x
U
a be 0.4332 standard (i) rea between 0.0332. deviation 10 0.4332
 W
60 When and of 10. 72, total Assuming (ii) frequency between that is 50 the N, and distribution then

o
60, expected (iii)

rld
60
1.2
Area from 0 to 1.2 0.3849. Percentage of observations between 60 to 72 0.3849
100 38.49
[ From the table the area between 0 and 1.26 0.3849]. Hence, the area between 60 and 72 0.3849
100 38.49 .] (ii) Between 50 and 60:

1 x
100 34.13
(iii) Beyond 72: Since half area is 0.5, hence area for more than 72 0.5 0.3849 0.1151
Percentage of observations more than 72 0.1151
100 11.51
(iv) Between 70 and 80 :
70
1
0
60 10
Area between 1 and 0 0.3413
Percentage of observations 0.3413
5.5 is 0.332N. Let N 100, then expected frequency is 3.32 .
1.83
for X 5.5, 2 5.5
1.5
for
1

2 65
1.5
1.83
50

10 1
60 10 10
Area 0.3413
72
rea 0.4664
60 10

10 3

10 3

10 10

2 80 10 60
20 2
10
Area 0.4772
Area between 70 and 80 0.4772 0.3413 0.1359
Percentage of observations between 70 and 80 0.1359
 l l
A E AMPLE 11.17 Solution: the 0.5 In S.D. Let left a X normal of of the the Mean

JN
distribution. ordinate distribution of the at distribution, x 31 45 of is the 0.31, items

T U
Standard hence are under the Deviation area 45 and from 8 of the the are
 W o
ordinate over distribution. 64. at Find x Since the 45 mean and the Mean area and

to

rld
Since the area to the right of the ordinate at x 64 is 0.08, hence the area between Mean and the ordinate at x 64 0.5
0.08 0.42. Corresponding value of for this area from the table is 1.4. Hence
64.
On solving both equations, we get
10, x 50
Mean, x 50 and standard deviation
E AMPLE 11.18
The mean length of steel bars produced by a company is 10 meter and standard deviation 20 cm. 5000 bars are purchased
by a building contractor.How many of these bars expected to be shorter than 9.75 meter in length? For your answer assume
that the length of steel are normally distributed and use the following extract from table of areas from mean to distances ( X X
) from mean under the normal curve :
X
1.10 1.15 1.20 1.25 1.30 Area .3643 .3749 .3849 .3944 .4032
Solution:
Here, X 10,
.2 10
1.25
Area from 1.25 to mean 0.3944. Area tothe left of the ordinate 0.5000 0.3944 0.1056. The proportion of the bars
which are shorter than 9.75 meter 0.1056 and the number 0.1056
5000 528
0.31 0.19. The corresponding value of Z for this area from the table is 0.5. Hence,
45

0.2 X

X
9.75
,X 9.75
x
10.

x
1.4 64
0.5 45
x
x 1.4
100 13.59
0.5
E AMPLE 11.19
15000 students sat for an examination. The mean mar s was 49 and the distribution of mar s had a standard deviation of 6.
Assuming that the mar s were normally distributed what pro portion of students scored (a) more than 55 mar s, (b) more
than 70 mar s.
Solution:
X X E AMPLE 6 49 1
(a)
The proportion of the students who got mar s between 49 scale) 0.3413 Hence the proportion of the students getting
mar s more than (b)

l
X
6
Al Solution: As life deviation expected between a time result 0 of to of and
 J N
tests the burn 60 11.21 A Z Z) bulbs hours for on under .3907 1.23 20000 (a) was .

TU
electric bulbs manufactured by a normally distributed with an average on the basis

of this information estimate more than 2150 hours and (b) less than normal curve is given: 1.33 1.43 1.53 1.63 .4082 .4236

W o
.4370 .4484 and 55 55 0.5 0.3413 (i,e., 0 and 0.1587 1 on standard company it

was found that the of 2040 hours and a standard the number of bulbs that are 1960 hours. the area, A (Area
1.73 1.83 .4582 .4664
rld
49

21 3.6
6 Area corresponding to 3.5 is 0.4998. Proportion of students getting mar s more than 70 0.5
0.4998 0.0002 and the required number of students 0.0002
E AMPLE 11.20
The mar s obtained in some degree examination are normally distributed , 500 and
100 ; we have to pass 550 students out of 674 students who are appearing in the examination. What should be the minimum
mar s for passing the examination ?
Solution: corresponding The proportion to 0.8162
550 674 0.8162. From the normal table the value of
of successful students 9. From
X
90 500 410
Thus minimum passing mar s 410

X
, where X Maximum mar s for passing
and X 55, X 49,
0.50 0.3162 is 9. Hence the value of to the left of mean is
15000 3
500 100
X
6 55

X
X
.9
70

Here X 2040, 1960, N 20000

(a)

X
A 60 2040
110
 1.83
60
Area to the right of 1.83 0.5
0.4664 0.0336
Required Number 0.0336
20000 672
(b)
X
1.33
Area to the left of
1.33 0.5 0.4082 .0918
Required Number 0.9018
20000 1836

11.40 1000 light bulbs with a mean life is normally distributed with standard 90 days ? (ii) It is decided to replace between
replacement if not more than 11.41 In a distribution exactly normal are the mean and S.D. of the distribution 11.42 The

l
mean height of the 1000 5 inch.How many wor ers are expected 11.43 ation 11.44 bulbs of selected 11.45 in of If 5

l
inch. a) b) The have bulbs values might If a regiment an ordinate greater between the an How Mumbai
 J
average height be is as than many expected 60 follows of Municipal life and 70 500 students

N
inch. of 70 to : students one inch. fail thousand Corporation have X in Probability the mean height of

soldiers of 1000 can be expected at 1.15 0.1251.]

T
all are wor ers 10 life height to the 7 ? normally deviation of be should bulbs 120 of : above in the a
 U
days together expire distributed steel items 20 72 are days. installed what (i) How interval in a

W
many new interval factory. will before replacement . are under 35 and 89 plant is 67 inch with a

inch in that plant?

with mean 65 inch installed 2000 bulbs in the streets burning hours in with a S.D. of 200 the first 700 burning hours ? the
table of 1 1.25 1.50 0.159 0.106 0.067 is 68.22 inch with a variance of 10.8 inch, to be over 6 feet all ? [ In a normal

o
distribution Their length of expire in less than should be allowed

are under 63. What

standard deviation of
and standard devi
of Mumbai. If these hours, what numbers the normal curve at
how many soldiers are to the right

rld
11.46 The income per month of a group of 10000 persons was found to be normally distributed with mean Rs. 750 with a
standard deviation of Rs. 50. show that of this group 95 had income
X

X
2150

X
1960

2040 60
exceeding Rs. 668 and only 5 had income exceeding Rs. 832. What was the lowest income among the richest 100?
11.47 In a statistical survey of 1000 small business firms in a city , it was found that their monthly average sales amounted to
Rs. 8000 with a standard deviation of Rs. 2000. Assuming that the sales are normally distributed,estimate :
a) the number of firms whose monthly average sales were less than Rs. 6000 and b) the number of firms whose monthly
average sales were between Rs. 7000 and Rs. 9000.
area under the normal curve

x A 0.1 0.5 1.0 1.5 2.0 Area 0.0398 0.1915 0.3413 11.48 A sales-tax officer has reported that the

average sales with during a year amount to Rs. 360000 with a standard deviation sales in these business are normally
distributed, find: a) the number of business the sales of which are over b) The percentage of business, the sales of which
are and Rs. 400000. c) The probability that the sales of business selected Area under the Normal Curve
 l l
X 0.25 0.40 0.50 11.49 items normal 1.2243 have distribution In is values a 0.39). normal

J N
under the distribution area 62. Area between Find 7 the 0.0987 percent means men

T U
and of and 0.1554 the standard x items 1.4757 0.1915 have deviation is Rs. of at
 W
0.4332 li ely the random 400000. 0.2257 of 0.60 500 to Rs. range 0.4772 business

o
100000.Assuming will be between over that he Rs. Rs. has 300000.

300000 that to deal the

values under 35 and 89 percent of the of the distribution (In a standard 0.43 and the area between mean and x

rld
Jointly Distributed Random Variables
Two Discrete Random Variables

ll
A DEFINITI N

J
NTU Wo

rld

Example 5.1
DEFINITI N
 ll
A
JN

T U
W

r
 ld

Example 5.2

(Example 5.1
continued)

Two Continuous Random Variables

 lJ
DEFINITI N
Al
N

U
 W

r
ld

( , )

Example 5.3

Surface ( )
Figure 5.1

)
volume under density surface above

Shaded rectangle

ll J
DEFINITI N A Example (Example continued 5.3 5.4
NTU

World
Example 5.5
Figure 5.2 Region of positive density for Example 5.5

All
J

T
 U World
for Example 5.5

1.5

0
0 .5 1 Figure 5.3 Computing

Shaded region

.5
(0, 1)
( ,1

,
)

)
(1, 0)
Independent Random

l
Al DEFINITI N
JN T

U
Variables

W
 or l

Example 5.6
Example 5.7 (Example 5.5 continued)