0% found this document useful (0 votes)
35 views50 pages

P&S Unit 1

This document covers the fundamentals of random variables and probability distributions, including definitions, types, and applications. It explains key concepts such as sample space, probability mass functions, and the calculation of expectation, mean, variance, and standard deviation. Additionally, it provides examples to illustrate the application of these concepts in various scenarios.

Uploaded by

bojjatrilok
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
35 views50 pages

P&S Unit 1

This document covers the fundamentals of random variables and probability distributions, including definitions, types, and applications. It explains key concepts such as sample space, probability mass functions, and the calculation of expectation, mean, variance, and standard deviation. Additionally, it provides examples to illustrate the application of these concepts in various scenarios.

Uploaded by

bojjatrilok
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 50

MATHEMATICS IV Cluster- I

Random Variables& Probability distributions

Introduction: In this Chapter we will discuss about basics of probability, random variables , distributions and some
important characterstics like mean, variance and standard deviation of distribution. Also we will discuss some important
probability distributions like Normal distributions.

Applications of random variables:

The no. of calls received at a telephone exchange during a definite time intervals is a random variable.

The no. of cosmic ray particles striking some area on the earth’s surface per until time depends on a multitude of random
factors.

Accept Sampling in Statistical Quality Control

Electronic Testing in Manufacturing goods

In the applications of theory of probability, it is more convenient to work numerical outcomes rather than non -
numerical outcomes because of this we introduce the idea of random variables.

The probability of happening of an event means that the amount of likelihood of the
happening of that event.
Outcome: The end result of an experiment
Random Experiment: An experiment which can be repeated any number of times under
identical conditions and the result of a particular trial is uncertain even though all possible
outcomes of that experiment are known in advance is called a random experiment. The
outcomes of random experiment are known as elementary events and a set of elementary
events is called as an event.
Examples: 1. Tossing of a coin
2. Throwing a die.
Sample space(Probability space): The set of all outcomes of a random experiment is called
sample space and it is denoted by S.
EX: In Tossing of two coins, S = { (H,H), (H,T), (T,H), (T, T)}
A random event is an outcome or a set of outcomes of a random experiment.
Exhaustive events: A set of events is said to be exhaustive , if it includes all the possible
events.
Mutually exclusive events: A set of events is said to be mutually exclusive if it the occurrence
of one of them prevent the occurrence of the others.
Equally likely events: If one of the events can not be expected to happen in preference to
another.
Probability of an Event: Let E be an event in a random experiment which contains ‘n’ mutually
exclusive and equally likely elementary events . If ‘m’ elementary events are favourable to
the event E, then the probability of E is defined as
Number of favourable elementary events to E m
P(E) = =
Total number of elementary events n

Ex: In a single throw of two dice find the probability of throwing a sum 10 of two dice.
Ans: Random experiment: Throwing of two dice.
Sample space = {(1, 1), (1,2),……..(1,6), (2,1),(2,2),…..,(2,6), …….
……,(6,1),(6,2), …….(6,6)}
Event E: Getting sum of two as 10.
Favourable events to E are (5,5),(4,6),(6,4).
3 1
∴ P(E) = =
36 12

Axioms of Probability:
1. P(E)≥0 2. P(S)=1 3. If A ∩ B = ∅, then P(A ∪ B) = P(A) + P(B)
4. P(A ∪ B) = P(A) + P(B) − P(A ∩ B)
̅) = 1 − P(A)
5. P(A 6. If B ∁ A, then P(B) < P(A)

DEFINITION OF RANDOM VARIABLE: A real valued function defined on the sample space S
of a random experiment is called the random variable, we denote random variables by X,Y,Z
etc.,.
Random variable assigns a real number to every element of the sample space.
For example: In tossing a coin,
the sample space 𝑆 = {H, T} , whose elements are non numerical values.
We convert into numerical values by using random variable
i.e. if the outcome is head, X(H) = 0 and
if the outcome is tail, X(T) = 1
X : S→R
H → 0
( ) ( )
T → 1
That is
X:S→R Such that{ω ∈ S/X(ω) = x, x ∈ R}, Then X=X(ω) is a R.V.
Types of Random Variables:
There are two types of random variables:
(i) Discrete random variables (ii) Continuous random variables
Discrete Random variable:
A real valued function defined on the sample space S of a random experiment is called the
random variable. A random variable X is said to be discrete if, it takes only a finite no. of
discrete values ( integral values ) in an interval.
Example: The number of rooms in the houses of a township.
Eg: 1 The no. of arrivals at ATM counter.
2. The no. of printing mistakes per page.
3. The no. of students in the class room
4. The no. of soldiers participated in the war
5. The no. of cars produced by the company in a week
6. The no. of patient joined the hospital etc,.
7. Consider Random experiment: Throwing of two dice
S = {(1,1), (1,2),….(1,6),
.

(6,1), (6,2),…..,(6,6)}

Let 𝑋(𝑤)= sum of the components in w.

𝑋(1,1) = 2, 𝑋(1,2) = 3, … … . , 𝑋(6,6) = 12.

Range of 𝑋 = {2,3,4, … 12}


Here, random variable X is a discrete random variable

Continuous random variable:


A real valued function defined on the sample space S of a random experiment is called the
random variable. A random variable X is said to be a continuous random variable if it
assumes all possible values between the limits.
Eg ; 1 The height of a student in a college
2 The life time of electronic component ;
3 The voltage in semi-conductor diode.
4 The trouble shooting capability of an IC chip in a circuit
5 The daily consumption of electric power
Probability mass function (pmf) :
Let X be a discrete random variable. Then a real valued function defined as p(𝑥) = P(X = x)
is called as the probability mass function of the random variable X
Properties:
i) p(x) ≥ 0, ∀x, (non negativity)
ii) ∑x p(x) = 1 (normality)

Eg:1. Tossing of a coin


R.V X : 0 1
pmf p(𝑥) ∶ 1/2 1/2
2. Random experiment: Tossing of two coins
𝑆 = { (𝐻, 𝐻), (𝐻, 𝑇), (𝑇, 𝐻), (𝑇, 𝑇)}
Let X(w)= number of heads is w.
Range of X ={0,1,2}
𝑝(0) = 𝑃(𝑋 = 0) = 𝑃({(𝑇, 𝑇)}) = ¼.
2
𝑝(1) = 𝑃(𝑋 = 1) = 𝑃({𝑤 ∈ 𝑆/𝑋(𝑤) = 1} = 𝑃({(𝐻, 𝑇), (𝑇, 𝐻)} = 4

𝑝(2) = 𝑃(𝑋 = 2) = 𝑃({𝑤 ∈ 𝑆/𝑋(𝑤) = 2} = 𝑃({(𝐻, 𝐻)} = 1/4


Observe that p(x)≥0 for all x
1 2 1
∑ p(x) = p(0) + p(1) + p(2) = + + = 1
4 4 4

Note: (X = k) = {w ∈ S/X(w) = k}
(X ≤ k) = {w ∈ S/X(w) ≤ k}
(X ≥ k) = {w ∈ S/X(w) ≥ k}
Probability distribution function [ cumulative distribution function]
Let X be a one-dimensional random variable. The probability distribution
function defined by F X (𝑥) = P (X ≤ 𝑥) where 𝑥 €R.

Ex: Random experiment: Tossing of two coins


𝑆 = { (𝐻, 𝐻), (𝐻, 𝑇), (𝑇, 𝐻), (𝑇, 𝑇)}
Let 𝑋(𝑤) = number of heads is w.
Range of 𝑋 = {0,1,2}
2
𝑝(0) = ¼, 𝑝(1) = 4 , 𝑝(2) = ¼
𝐹𝑋 (0) = 𝑃(𝑋 ≤ 0) = 𝑃(𝑋 = 0) = 𝑝(0) = ¼
2 3
𝐹𝑋 (1) = 𝑃(𝑋 ≤ 1) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) = 𝑝(0) + 𝑝(1) = ¼ + 4 = 4
2
𝐹𝑋 (2) = 𝑃(𝑋 ≤ 2) = 𝑝(0) + 𝑝(1) + 𝑝(2) = ¼ + 4 + ¼ = 1

Probability Distribution of Discrete random variable.


Let X be a discrete random variable which takes finite values 𝑥1 , 𝑥2 , … , 𝑥𝑛 .

Let us suppose that p1, p2,… , pn are the respective probabilities

i.e p(𝑥i ) = pi the set of possible values of 𝑥 together with their respective probabilities is
called the probability distribution of X.
The probability distribution of X is represented as follows.
X=𝑥 𝑥1 𝑥2 𝑥3 ………… 𝑥𝑛
p(x) p1 p2 p3 ………… pn

Ex: Random experiment: Tossing of two coins


𝑆 = { (𝐻, 𝐻), (𝐻, 𝑇), (𝑇, 𝐻), (𝑇, 𝑇)}
Let 𝑋(𝑤) = number of heads is w.
Range of 𝑋 = {0,1,2}
𝑝(0) = ¼, 𝑝(1) = 2/4, 𝑝(2) = ¼
The probability distribution of X is
X=𝑥 0 1 2
p(x) ¼ 2/4 ¼

Ex:2 Throwing of two dice.


𝑆 = {(1,1), (1,2), … . (1,6),
……………….. ‘
(6,1), (6,2), … . . , (6,6)}

Let 𝑋(𝑤) = sum of the numbers in w.

𝑋(1,1) = 2, 𝑋(1,2) = 3, … … . , 𝑋(6,6) = 12.


Range of 𝑋 = {2,3,4,5,6,7,8,9,10,11,12}
1
Clearly p(2) = P(X = 2) = P({(1,1)} = 36
2
p(3) = P(X = 3) = P({(1,2), (2,1)} = 36
3
p(4) = P(X = 4) = P({(1,3), (2,2), (3,1)} = 36 .

1
p(12) = P(X = 12) = P({(6,6)} = 36

The probability distribution of X is


X=𝑥 2 3 4 5 6 7 8 9 10 11 12
p(𝑥) 1 2 3 4 5 6 5 4 3 2 1
36 36 36 36 36 36 36 36 36 36 36

Expectation, Mean, Variance and Standard deviation of a discrete probability distribution


Let X be a discrete random variable which takes finite values x1 , x2 , … , xn with respect to
probabilities p1, p2,… , pn .

The probability distribution of X is


X=𝑥 𝑥1 𝑥2 𝑥3 ………… 𝑥𝑛
p(𝑥) 𝑝1 𝑝2 𝑝3 ………… 𝑝𝑛

Expectation:
The Expectation of X is denoted by E(X) and defined as
E(X) = ∑ni=1 pi 𝑥i = 𝑝1 𝑥1 + 𝑝2 𝑥2 + ⋯ … . +𝑝𝑛 𝑥𝑛
E(X 2 ) = ∑ni=1 pi 𝑥𝑖 2 , E(X 3 ) = ∑ni=1 𝑝𝑖 𝑥𝑖 3
In general E(g(X)) = ∑ni=1 pi g(𝑥i )
Properties:
1. If X discrete random variable and k is a constant , then E(X + k) = E(X) + k
Sol: E(X + k) = ∑𝑥(𝑥 + k). p(𝑥)
= ∑𝑥[kp(𝑥) + 𝑥𝑝(𝑥)]
= k ∑𝑥 p(𝑥) + ∑𝑥 𝑥𝑝(𝑥)
= k ∑𝑥 p(𝑥) + ∑𝑥 𝑥 𝑝(𝑥)
= k. 1 + E(X) = E(X) + k
2. If X is discrete random variable and α, β are constants , then E(α + βX) = α + βE(X)
Sol: E(α + βX) = ∑x(α + βx). p(x)
= ∑x[α p(x) + β. xp(x)]
= ∑x α p(x) + ∑x βxp(x)
= α ∑x p(x) + β ∑x x p(x)
= α. 1 + β. E(X)
3. If X,Y are two discrete random variables , then
E(αX + βY) = α E(X) + βE(Y)

E(X + Y) = E(X) + E(Y)


4. If X,Y are two independent random variables , then
E(XY) = E(X)E(Y)
MEAN:
∑ pi 𝑥𝑖
The mean of the distribution is denoted by µ and defined as µ= ∑ pi
= ∑ pi 𝑥𝑖 = E(X)

In probability theory µ=E(X)


Variance : The arithmetic mean of the sum of the squares of the deviation taken from their
mean is defined as Variance

The variance of the distribution is denoted by σ2


σ2 = V(X) = E (X − μ)2
= E(X 2 + μ2 − 2μX)
= E(X 2 ) + E(μ2 ) − 2μE(X)

= E(X 2 ) + μ2 − 2μ. μ
= E(X 2 ) + μ2 − 2μ2
σ2 = E(X 2 ) − μ2 =E(X 2 ) − E(X)2

Properties:
(1) V(k) = 0
2
V(k) = E(k 2 ) − E(k)2 = ∑ pi k 2 − (∑ pi k)
= k 2 ∑ pi − (k ∑ pi )2

=k 2 − k 2 = 0
2
(2) V(kX) = E((kX)2 ) − (E(kX))
2
= ∑ pi (kxi )2 − (∑ pi kxi )

= k 2 ∑ pi xi 2 − (k ∑ pi xi )2

=k 2 E(X 2 ) − k 2 E(X)2 = k 2 (E(X 2 ) − E(X)2 ) = k 2 V(X)


2
(3) V(X + k) = E((X + k)2 ) − (E(X + k))
2
= ∑ pi (k + xi )2 − (∑ pi (k+xi ))

2
= ∑ pi (k 2 + xi 2 + 2kxi ) − (∑ pi xi + pi k)

= ∑ pi xi 2 + 2k ∑ pi xi + k 2 ∑ pi − (∑ pi xi + k ∑ pi )2

= ∑ pi xi 2 + 2k ∑ pi xi + k 2 − (∑ pi xi + k )2

=E(X 2 ) + 2kE(X) + k 2 − (k 2 + 2kE(X) + E(X)2


= E(X 2 ) − E(X)2=V(X)

(4) V(aX + b) = V(aX) = a2 V(X)

(5) V(X ± Y) = V(X) ± V(Y)


Standard Deviation:
The standard deviation of X is denoted by σ and is defined as

σ = √V(X)

Thus σ = √V(X) = √E (X − μ)2 = √E(X 2 ) − E(X)2

Problems:

Prob1 : The random variable X, representing the number of errors per 100
lines of software code, has the following probability mass function

RV X: 0 1 2 3 4 5 6 7
pmf p(x): 0 k 2k 2k 3k 𝒌𝟐 2𝒌𝟐 7𝒌𝟐 + 𝒌

Find Constant k, P(X<6), P(0<X<5), Mean ,Variance and standard deviation , cdf

Solution

From the property of normality ∑𝑛𝑖=0 𝑝(𝑥𝑖 ) = 1


∑7𝑖=0 𝑝(𝑥𝑖 ) = 1

⇒ 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + − − − − − + 𝑃(𝑋 = 7) = 1

⇒ 0 + 𝐾 + 2𝐾 + 2𝐾 + 3𝐾 + 𝐾 2 + 2𝐾 2 + (7𝐾 2 + 𝐾) = 1

⇒ 7𝐾 2 + 9𝐾 − 1 = 0

(10𝐾 − 1)(𝐾 + 1) = 0,

1
𝐾= 𝑜𝑟 𝐾 = −1
10
1
∴𝐾= , 𝑝(𝑥)𝑖𝑠 𝑛𝑜𝑛 − 𝑛𝑒𝑔.
10

The probability mass function of X is

RV X: 0 1 2 3 4 5 6 7
1 2 2 3 1 2 17
pmf p(𝑥): 0 10 10 10 10 100 100 100

(ii) 𝑃(𝑋 < 6) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) +


𝑃(𝑋 = 5)

= 𝑝(0) + 𝑝(1) + − − − − − + 𝑝(5)


1 2 2 3 1 81
=0 + + + + + = 100.
10 10 10 10 100

𝑃(0 < 𝑋 < 5) = 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4)

= 𝑝(1) + 𝑝(2) + 𝑝(3) + 𝑝(4)


1 2 2 4
= 0 + 10 + + = 5.
10 10

(iii) 𝑀𝑒𝑎𝑛 𝜇 = 𝐸(𝑋) = ∑𝑛𝑖=0 𝑥𝑖 𝑝(𝑥𝑖 )


𝜇 = ∑7𝑖=0 𝑥𝑖 𝑝(𝑥𝑖 )

= 0. 𝑝(0) + 1. 𝑝(1)+ . . +7. 𝑝(7)


1 2 2 3 1 2 17
= 0.0+ 1 . 10 + 2. + 3. + 4. 10 + 5. +6 . 100 + 7 . 100
10 10 100

1 4 6 12 5 12 119
= + + + + + +
10 10 10 10 100 100 100

366
𝑀𝑒𝑎𝑛 𝜇 = 100

Variance: 𝝈𝟐 = 𝑉(𝑋) = 𝐸(𝑋 2 ) − 𝑚𝑒𝑎𝑛2

E(𝑋 2 ) = ∑𝑛𝑖=0 𝑥𝑖 2 𝑝(𝑥𝑖 )

= 02 𝑝(0) + 12 𝑝(1) + − − − − 72 𝑝(7)


1 2 2 3 1 2 17
= 0.0 + 1 . 10 + 4. + 9. + 16. 10 + 25. +36 . 100 + 49 . 100
10 10 100

1 8 18 48 25 72 833
= + + + + + +
10 10 10 10 100 100 100

1680
= 100

1680 366 2
𝑉(𝑋) = 𝐸(𝑋 2 ) − 𝑚𝑒𝑎𝑛2 = − (100)
100

34044
= = 3.4044
10000

Standard deviation, 𝑆𝐷 (𝑋), 𝜎 = √𝑉(𝑋) =√3.404 = 1.84

CDF, F(x)= P (X≤ 𝒙)

The probability distriution of X is

RV X: 0 1 2 3 4 5 6 7
1 2 2 3 1 2 17
PMF p(𝑥): 0 10 10 10 10 100 100 100

1 3 5 8 81 83 100
CDF F X (𝑥): 0 10 10 10 10 100 100 100
Prob 2: Two dice are thrown. Let X assign to each point (a,b) in the sample space S , the
maximum of its numbers i.e X(a,b)= max(a,b). Find the probability distribution. Also find
mean and variance of the distribution.
Sol: Random experiment: Throwing of two dice
S = {(1,1), (1,2),….(1,6),
.

(6,1), (6,2),…..,(6,6)}

Let X(a,b)=max{a,b} .

𝑋(1,1) = 1, 𝑋(1,2) = 𝑋(2,1) = 2,


𝑋(1,3) = 𝑋(3,1) = 𝑋(2,3) = 𝑋(3,2) = 3, 𝑋(4,1
𝑋(1,4) = 𝑋(4,1) = 𝑋(2,4) = 𝑋(4,2) = 𝑋(3,4) = 𝑋(4,3) = 4
𝑋(1,5)(1,5)(1,5)(1,5)(1,5)(1,5) = ⋯ … … … … … … … = 𝑋(5,4) = 5
𝑋(6,1) = ⋯ … … … … … … … … … … … … … = 𝑋(5,6) = 6.
Range of 𝑋 = {1,2,3,4,5,6}
1
𝑝(1) = 𝑃(𝑋 = 1) = 36
3
𝑝(2) = 𝑃(𝑋 = 2) = 𝑃{(1,2), (2,1), (2,2)} = 36
….
Range of X = {2,3,4,…12}

The probability distribution of X is

RV X: 1 2 3 4 5 6
1 3 5 7 9 11
pmf p(𝑥): 36 36 36 36 36 36

𝑀𝑒𝑎𝑛 𝜇 = 𝐸(𝑋) = ∑𝑛𝑖=0 𝑥𝑖 𝑝(𝑥𝑖 )

𝜇 = ∑6𝑖=1 𝑥𝑖 𝑝(𝑥𝑖 ) = 1. 𝑝(1)+ . . +6. 𝑝(6)


1 3 5 7 9 11 161
= 1 .36 + 2. 36 + 3. 36 + 4.36 + 5. 36 +6 .36 = 36 = 4.47

𝑉(𝑋) = 𝐸(𝑋 2 ) − (𝐸(𝑋))2 =∑ 𝑝𝑖 𝑥𝑖 2 − 𝜇 2


1 3 5 7 9 11
= 1 .36 + 4. 36 + 9. 36 + 16.36 + 25. 36 +36 .36 = 4.472
791
= − 19.9808 = 21.97 − 19.9808 = 1.9912
36

Standard deviation, 𝑆𝐷 (𝑋), 𝜎 = √𝑉(𝑋) =√1.9912 = 1.4111


Prob 3 Let the random variable X represent the number of defective
parts for a machine when 6 parts are sampled form a production line and
tested. The following is the probability distribution of

X: 0 1 2 3 4 5 6

p(𝒙): k 3k 5k 7k 9k 11k 13k

Determine the value of k.

1 Find 𝑷(𝑿 < 𝟒), 2 𝑷(𝑿 ≥ 𝟓), 𝒂𝒏𝒅 3 𝑷(𝟑 < 𝑿 ≤ 𝟔).

What will be the minimum value of k so that P (𝑿 ≤ 𝟐) > 𝟎. 𝟑.

Solution: 𝑷(𝑿 = 𝒙) = 𝒑(𝒙)

Form normality, ∑𝑛𝑖=0 𝑥𝑖 𝑝(𝑥𝑖 )

⇒ ∑6𝑖=0 𝑥𝑖 𝑝(𝑥𝑖 ) = 1

⇒ k + 3k + 5k +------- 13k = 1
1
K= 49

1 𝑃(𝑋 < 4) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3)


1 3 5 7 16
= 𝑃(0) + 𝑝(1) + 𝑝(2) + 𝑝(3) = + + + = 49
49 49 49 49

11 13 24
2 𝑃(𝑋 ≥ 5) = 𝑃(𝑋 = 5) + 𝑃(𝑥 = 6) = 𝑝(5) + 𝑝(6) = + = 49
49 49

9 11 13 33
3 𝑃(3 < 𝑋 ≤ 6) = 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) + 𝑃(𝑋 = 6) = + + = 49
49 49 49

4 𝑃(𝑋 ≤ 2) > 0.3 ⇒ 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) > 0.3

⇒ 𝑝(0) + 𝑝(1) + 𝑝(2) > 0.3

⇒ k + 3k + 5k > 0.3 ⇒ 9k > 0.3


1
⇒k > 30

1
the min. Value of k = 30

Prob 4:For the following probability distribution find (i) E(X) (ii)E(X2) (iii) E(2X+1)2
X=x -2 7 12
𝟏 𝟏 𝟏
p(x): 𝟔 𝟐 𝟑
1 1 1 43
Sol : E(X) = ∑ pi 𝑥𝑖 = (-2) .6 + 7. 2 + 12. 3 = 6

𝐸(𝑋 2 ) =∑ 𝑝𝑖 𝑥𝑖 2
1 1 1 439
= 4 .6 + 49. 2 + 144. 3 = 6

E((2X + 1)2 ) = E(4X 2 + 4X + 1)


= 4 E(X 2 )+4E(X) + E(1)
439 43 967
= 4. +4. 6 +1=
6 3

Prob 5: From a lot of 10 items containing 3 defectives, a sample of four items is drawn at
random. If the discrete random variable X denotes the number of defective items in the
sample, write the probability distribution when the sample is drawn without replacement.
Sol: Total number of items =10
Defective items = 3
Random experiment: Collection of sample of size 4
Random variable X is the number of defective items in sample.
Clearly range of X ={0,1,2,3}
Total number of samples is 10C4

Zero defective:
Favourable number of cases( number of samples with zero defectives and 4 non
defective) is 7C4

p(0)=P(X=0)=Probability of no defectives
7C 7! 4!6! 4×5×6 1
= 10 4 = 4!3! × = 8×9×10 = 6
C4 10!

One defective:
Favourable number of cases( number of samples with 1 defectives and 3 non
defective) is 3C1 × 7C3

p(1)=P(X=1)=Probability of one defective and 3 non defectives


3C1 ×7C3 7! 4!6! 4×5×6 1
= = 3 × 3!4! × = 3 × 8×9×10 = 2
10C4 10!

Two defective:
Favourable number of cases( number of samples with 2 defectives and 2 non
defective) is 3C2 × 7C3

p(2)=P(X=2)=Probability of two defectives and 2 non defectives


3C2 ×7C2 7! 4!6! 3×4×6 3
= = 3 × 2!5! × = 3 × 8×9×10 = 10
10C4 10!

Three defective:
Favourable number of cases( number of samples with 3 defectives and 1 non
defective) is 3C3 × 7C1

p(3)=P(X=3)=Probability of three defectives and one non defective


3C3 ×7C1 4!6! 2×3×4 1
= =1×7× = 7 × 7×8×9×10 = 30
10C4 10!

Probability Distribution:
X=x 0 1 2 3
p(x) 1 1 3 1
6 2 10 30

Prob 6: A player tosses three coins. He wins 400 Rs if 3 heads, 200 Rs if 2 heads 100 Rs if 1
head occurs. On the other hand, he looses 1500 Rs if 3 tails occur. Find the expected gain of
the player.
Sol: Random experiment: Tossing of three coins.
Sample Space
𝑆 = {(𝐻, 𝐻, 𝐻), (𝐻, 𝐻, 𝑇), (𝐻, 𝑇, 𝐻), (𝑇, 𝐻, 𝐻), (𝐻, 𝑇, 𝑇), (𝑇, 𝐻, 𝑇), (𝑇, 𝑇, 𝐻), (𝑇, 𝑇, 𝑇)}
Number of elements in S is 𝑛 = 8.
Let Random variable X be the gain of the player.
Range of 𝑋 = {400, 200, 100, −1500}
1
Probability of gain 400 is 𝑝(400) = 𝑃(𝑋 = 400) = 𝑃{(𝐻, 𝐻, 𝐻)} = 8
3
𝑝(200) = 𝑃(𝑋 = 200) = 𝑃{(𝐻, 𝐻, 𝑇), (𝐻, 𝑇, 𝐻), (𝑇, 𝐻, 𝐻)} = 8
3
𝑝(100) = 𝑃(𝑋 = 100) = 𝑃{(𝐻, 𝑇, 𝑇), (𝑇, 𝐻, 𝑇), (𝑇, 𝑇, 𝐻)} = 8
1
𝑝(−1500) = 𝑃(𝑋 = −1500) = 𝑃{(𝑇, 𝑇, 𝑇)} = 8

The Probability distribution of X is


X=x 400 200 100 -1500
p(x) 1 3 3 1
8 8 8 8

E(X) = ∑ pi xi

1 3 3 1 200
Expected gain E(X) = 400 .8 + 200. 8 + 100. 8 -1500. 8=- = −25
8

Thus expected gain by the player is ‘ -25 ‘ .

Prob 7: A fair die is rolled . Obtain the expectation of the number on the die.
Sol: Random experiment: Throwing a die.
Sample Space 𝑆 = {1,2,3,4,5,6}
𝑋(𝑤) = number on the face of the die.
Range of 𝑋 = {1,2,3,4,5,6}
The probability distribution is
X=x 1 2 3 4 5 6
p(x) 1 1 1 1 1 1
6 6 6 6 6 6

1 1 1 1 1 1 21
E(X) = ∑ pi xi =1 .6 + 2. 6 + 3. 6 +4 .6 + 5. 6 + 6. 6= = 3.5
6

Prob 8: For the following probability distribution show that the missing probability is
0.778
𝐗=𝐱 -3 -2 -1 0 1 2 3
𝐩(𝐱) 0.001 0.01 0.1 ? 0.1 0.01 0.001

Sol: Let p(0) = k


We know that ∑ pi = 1
⟹ 0.001 + 0.01 + 0.1 + k + 0.1 + 0.01 + 0.001 = 1
⟹ k + 0.222 = 1
⟹ k = 0.778
Hence the missing probability p(0) = 0.778
𝐤
Prob 9: Given that 𝐟(𝐱) = is the probability mass function for a random variable that
𝟐𝐱
takes the values 𝐱= 1,2,3,4. Find (i) k (ii) mean&variance
Sol: The probability distribution of X is
X=x 1 2 3 4
p(x) k k k k
8
2 4 6

(i) We know that∑ pi = 1


k k k k 25 24
⟹ 2 + 4 + 6 + 8 = 1 ⟹ k 24 = 1 ⟹ k = 25
k k k k k k k k 48
(ii) Mean: μ = ∑ pi xi = 2 + 2. 4 + 3. 6 + 4. 8 = 2 + 2 + 2 + 2 = 2k = 25

(iii) Variance: 𝛔𝟐 = V(X) = E(X 2 ) − μ2

= ∑ pi x i 2 − μ2
𝑘 𝑘 𝑘 𝑘 48 2
=2 + 4. 4 + 9. 6 + 16. 8 − (25)

48 2 24 48 2
= 5k − (25) = − (25) = 1.1136
5

Practice problems:Tutorial
4.

Continuous random variable:


A random variable X is said to be a continuous random variable if it assumes all possible
values in an interval of a domain.
Eg ; 1 Height of student in a class.
2 The life time of electronic component ;
3 The voltage in semi-conductor diode.
4 The trouble shooting capability of an IC chip in a circuit
5 The daily consumption of electric power
Probability density function (pdf) :
Let X is a continuous random variable. Let f(x) be the continuous function, so that f(x)dx
dx dx
represents the probability that the variable X falls into the interval [x − x+ ].
2 2
dx dx
Symbolically it can be expressed as P [x − ≤X≤x+ ] = f(x)dx.
2 2

Then the function f(x) is called as the probability density function of X. The curve y = f(x) is
called the probability curve.
If f(x) is density function of a continuous random variable X, then the probability of X falls in
to the interval (a, b) is given as
b
P(a < X < b) = ∫a f(x)dx = area under the probability curve y = f(x) between X = a and
X = b.
Properties of density function:
1 f(x) ≥ 0 (non negativity)

2 ∫−∞ f(x)dx = 1 (normality)

3. In continuous distribution inclusion or non-inclusion of end points does not


change the probability.
i.e P(a ≤ X ≤ b) = P(a < X ≤ b) = P(a ≤ X < b) = P(a < X < b)

Example: A function f(x) is defined as follows:


0, x < 2,
1
f(x) = { 18 (2x + 3), 2 ≤ x ≤ 4 , Show that it is a p.d.f.
0, x>4
Sol: Non negativity f(x) ≥ 0
Now,
1 11
If x = 4, f(4) = (2(4) + 3) = >0
18 18
1 7
If x = 2, f(2) = (2(2) + 3) = >0
18 18

f(x) ≥ 0

Normality : ∫−∞ f(x)dx = 1


2 4 ∞
∫ f(x)dx = ∫ f(x)dx + ∫ f(x)dx + ∫ f(x)dx
−∞ 2 4
−∞
2
1 4 ∞
= ∫ (0)dx + ∫ (2x + 3)dx + ∫ (0)dx
−∞ 18 2 4
4
1 x2
= [2. + 3x]
18 2 2
1 2
= [x + 3x]42
18
1
= [28 − 10]
18
= 1.

Since f(x) ≥ 0 , ∀x & ∫−∞ f(x)dx = 1, given f(x)is a pdf.

Problem: A function f(x) is defined as follows:


1
(3 + x)2 , −3 ≤ x < −1,
16
1
f(x) = (6 − 2x 2 ), −1 ≤ x < 1 ,
16
1 2
{ 16 (3 − x) , 1 ≤ x ≤ 3
Show that it is a p.d.f.
Solution : f(x) is clearly ≥ 0 for every x in [−3,3]and
∞ −1 1 3
∫−∞ f(x)dx = ∫−3 f(x)dx + ∫−1 f(x)dx + ∫1 f(x)dx
−1 1 1 1 3 1
= ∫−3 (3 + x)2 dx + ∫−1 (6 − 2x 2 )dx + ∫1 (3 − x)2 dx
16 16 16

1 (3+x)3 −1 1 1 1 (3−x)3 3
= 16 . [ ]−3 + 16 . 2 ∫0 (6 − 2x 2 )dx + 16 . [ ]
3 −3 1
1
1 1 2x3 1
= 48 (8 − 0) + 8 (6x − ) − 48 (0 − 8)
3 0

1 1 2 1 1 1 1 1 2
= 6 + 8 (6 − 3) + 6 = 3 + 4 (3 − 3) = 3 + 3 = 1

Cumulative distribution function(cdf):

The cumulative distribution function of a continuous random variable 𝑋 is


denoted 𝐹𝑋 (𝑥) and is defined by
𝑥
𝐹(𝑥) = 𝑃( 𝑋 ≤ 𝑥 )= ∫−∞ 𝑓(𝑥)𝑑𝑥 where 𝑥 ∈ ℝ.

Some Properties of a c.d.f.

1. 0 ≤ 𝐹𝑋 (𝑥) ≤ 1, −∞ < 𝑥 < ∞, 𝐹𝑋 (−∞)=0, 𝐹𝑋 (∞)=1

2. 𝐹 ′ (𝑥) = 𝑓(𝑥)

3. P(a< 𝑋 <b) =P(a ≤X<b)= P(a<X≤b)= P(a ≤ X ≤b)


𝑏
= ∫𝑎 𝑓(𝑥)𝑑𝑥 = F(b)–F(a)

Expectation, Mean, Median, Mode and variance for continuous probability distribution:
Let X be a continuous random variable and f(x) be its density function.
Expectation(Mean):
The mean of the distribution is denoted by μ and id defined as

μ = ∫−∞ xf(x)dx = E(X)

E(X 2 ) = ∫−∞ x 2 f(x)dx

E(X 3 ) = ∫−∞ x 3 f(x)dx

E(g(x)) = ∫−∞ g(x)f(x)dx

Median:
It is a point which divides the entire distribution in to two equal parts

We know that ∫−∞ f(x)dx = 1
M ∞
If M is the median then∫−∞ f(x)dx = ∫M f(x)dx = 1/2

Mode : Mode is a value of x where the density function f(x) has maximum.
So the mode x is given by f ′ (x) = 0 and f ′′ (x) < 0.
Variance: The variance is denoted by σ2 or V(X) and is defined by

σ2 = E[(X − μ)2] = ∫−∞(x − μ)2 f(x)dx

= ∫−∞[x 2 − 2μx + μ2 ]f(x)dx
∞ ∞ ∞
= ∫−∞ x 2 f(x)dx − 2μ ∫−∞ xf(x)dx + μ2 ∫−∞ f(x)dx

= E[X 2 ] − 2μE[X] + μ2

= E[X 2 ] − 2μ2 + μ2 = E[X 2 ] − μ2

Standard Deviation:
The standard deviation of X is denoted by σ and is defined as

σ = √V(X)

Thus σ = √V(X) = √E (X − μ)2


5. If X is continuous variable and Y = α + βX α, β are constants , then E(Y) = α +
βE(X)

Sol: E(Y) = E(α + βX) = ∫−∞(α + βx)f(x)dx
∞ ∞

= α ∫ f(x)dx + β ∫ xf(x)dx
∞ ∞

= α. 1 + β. E(X) = α + βE(X)
2
V(Y) = E((α + βX)2 ) − (E(α + βX))

= E(α2 + 2αβX + β2 X 2 ) − (E(α) + βE(X))2


2 2
= E(α2 )+2αβE(X) + β2 E(X 2 ) − (E(α)) − β2 (E(X)) − 2αβE(X)
2
= β2 (E(X 2 )-(E(X)) ) = β2 V(X)

Problem 1

Suppose that the error in the reaction temperature for a controlled laboratory
experiment is a continuous random variable X having the probability density
function

𝒄𝒙 (𝟐 − 𝒙), 𝒊𝒇 𝟎 ≤ 𝒙 ≤ 𝟐,
𝒇(𝒙) = { .
𝟎, 𝒐𝒕𝒉𝒆𝒓𝒘𝒊𝒔𝒆

where c is a constant. Find (i) c (ii) 𝑃(𝑋 ≤ 1)(iii) mean (iv) variance.

Solution :

Since the total probability is unity (normality), we have



∫ 𝑓(𝑥)𝑑𝑥 = 1.
−∞

0 2 ∞
⟹ ∫ 𝑓(𝑥)𝑑𝑥 + ∫ 𝑓(𝑥)𝑑𝑥 + ∫ 𝑓(𝑥)𝑑𝑥 = 1
−∞ 0 2

2
⟹ ∫ 𝑐𝑥(2 − 𝑥)𝑑𝑥 = 1
0

2
𝑥3 2
⟹ 𝑐 (𝑥 − ) = 1
3 0
8
⟹c(4 − 3) = 1

4𝑐
⟹ =1
3
⟹ 𝒄 = 𝟑/𝟒

The pdf of 𝑿 is

3𝑥
(2 − 𝑥), 𝑖𝑓 0 ≤ 𝑥 ≤ 2
𝑓(𝑥) = { 4
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

(ii) 𝑃(𝑋 ≤ 1),


1
𝑃(𝑋 ≤ 1) = ∫−∞ 𝑓(𝑥)𝑑𝑥

0 1
= ∫ f(x)dx + ∫ f(x)dx
−∞ 0
0 1 3x
= ∫−∞(0) dx + ∫0 (2 − x)dx
4
3 1
= ∫ x(2 − x), dx
4 0
3 1
= ∫ (2x − x 2 )dx
4 0
1
3 2x2 x3
=4 ( − )
2 3 0
1
3 x3
= ( x2 − )
4 3 0

3 1
= [(1 − 0) − (3 − 0)]
4

3 1 3 2 1
=4 ( 1− ) =4. =
3 3 2


Mean of X = 𝐸(𝑋) = ∫−∞ 𝑥𝑓(𝑥)𝑑𝑥

2 3𝑥
𝜇 = ∫0 𝑥. (2 − 𝑥)𝑑𝑥
4

3 2
= 4 ∫0 (2𝑥 2 − 𝑥 3 )𝑑𝑥

2
3 2𝑥 3 𝑥4
= 4( − )
3 4 0

3 24 24
= 4(3 − )
4
3 1 1
= 4 (24 ) (3 − 4)

12
= 12 = 1

Variance:
By def. 𝑉(𝑋) = 𝜎 2 = 𝐸(𝑋 2 ) − 𝜇 2

𝑉(𝑋) = ∫−∞ 𝑥 2 𝑓(𝑥)𝑑𝑥 − 𝜇 2

2 3𝑥
= ∫0 𝑥 2 . (2 − 𝑥)𝑑𝑥 − 𝜇 2
4

3 2
= 4 ∫0 (2𝑥 3 − 𝑥 4 )𝑑𝑥 − 𝜇 2

2
3 𝑥4 𝑥5
= 4 (2. − ) − 𝜇2
4 5 0

3 32 32
= 4[4 − ] − 𝜇2
5

1 1
= 24 (4 − 5) − 𝜇 2

6
= 5 − 𝜇2

6 1
𝑉(𝑋) = 𝜎 2 = 𝐸(𝑋 2 ) − 𝜇 2 = − 12 = 5
5

6. The shelf life in days, for bottles of a certain prescribed medicine is a


random variable 𝑿 having the density function
𝒙
, 𝒊𝒇 𝟎 < 𝒙 ≤ 𝟏,
𝟐
𝟏
, 𝒊𝒇 𝟏 < 𝒙 ≤ 𝟐,
𝒇(𝒙) = 𝟐
𝟑−𝒙
, 𝒊𝒇 𝟐 < 𝒙 < 𝟑,
𝟐
{ 𝟎, 𝒆𝒍𝒔𝒆𝒘𝒉𝒆𝒓𝒆

Find expected value 𝒐𝒇 𝑮(𝑿) = 𝑿𝟐 − 𝟓𝑿 + 𝟑.

Solution :

The expectation of any function 𝐺(𝑋)𝑖𝑠 𝑔𝑖𝑣𝑒𝑛 𝑏𝑦



𝐸[𝐺(𝑋)] = ∫ 𝐺(𝑥)𝑓(𝑥)𝑑𝑥
−∞

∴ 𝐸(𝑋 2 − 5𝑋 + 3) = ∫ (𝑥 2 − 5𝑥 + 3))𝑓(𝑥)𝑑𝑥
−∞

1 2 3
𝑥 1 3−𝑥
= ∫ (𝑥 2 − 5𝑥 + 3). 𝑑𝑥 + ∫ (𝑥 2 − 5𝑥 + 3) . 𝑑𝑥 + ∫ (𝑥 2 − 5𝑥 + 3). ( ) 𝑑𝑥
0 2 1 2 2 2

1 1 3 2
1 2 2 1 3
= ∫ (𝑥 − 5𝑥 + 3𝑥)𝑑𝑥 + ∫ (𝑥 − 5𝑥 + 3)𝑑𝑥 + ∫ (−𝑥 3 + 8𝑥 2 − 18𝑥 + 9)𝑑𝑥
2 0 2 1 2 2
1 2 3
1 𝑥 4 5𝑥 3 3𝑥 2 1 𝑥 3 5𝑥 2 1 𝑥 4 8𝑥 3 18𝑥 2
= ( − + ) + ( − + 3𝑥) + (− + − + 9𝑥)
2 4 3 2 0 2 3 2 1
2 4 3 2 2

1 13 19 44 11
= 24 − 12 − 24 = − 24 = − 6

11
Alternatively 𝐸(𝑋 2 − 5𝑋 + 3) = 𝐸(𝑋 2 ) + 5𝐸(𝑋) + 3=− 6

3 A function f(x) is defined of x as follows:

𝟏
(𝟑 + 𝒙)𝟐 , −𝟑 ≤ 𝒙 < −𝟏,
𝟏𝟔
𝟏
𝒇(𝒙) = (𝟔 − 𝟐𝒙𝟐 ), −𝟏 ≤ 𝒙 < 𝟏 ,
𝟏𝟔
𝟏 𝟐
{ 𝟏𝟔 (𝟑 − 𝒙) , 𝟏 ≤ 𝒙 ≤ 𝟑

find the Mean and variance of X

Solution

Mean of X 𝑓(𝑥) = ∫−∞ 𝑥𝑓(𝑥)𝑑𝑥

1 −1 1 1 1 3
= 16 ∫−3 𝑥(3 + 𝑥))2 𝑑𝑥 + 16 ∫−1 𝑥(6 − 2𝑥 2 )𝑑𝑥 + 16 ∫1 𝑥(3 − 𝑥)2 𝑑𝑥

1 −1 1 3
= 16 ∫−3 𝑥(9 + 𝑥 2 + 6𝑥)𝑑𝑥 + 0 + 16 ∫1 𝑥(9 − 6𝑥 + 𝑥 2 )𝑑𝑥

[∵ 𝑡ℎ𝑒 𝑖𝑛𝑡𝑒𝑔𝑟𝑎𝑛𝑑 𝑜𝑓 𝑡ℎ𝑒 𝑠𝑒𝑐𝑜𝑛𝑑 𝑖𝑛𝑡𝑒𝑔𝑟𝑎𝑙 𝑖𝑠 𝑜𝑑𝑑 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 ]

1 −1 1 3
= 16 ∫−3 (9𝑥 + 6𝑥 2 + 𝑥 3 )𝑑𝑥 + 16 ∫1 (9𝑥 − 6𝑥 2 + 𝑥 3 )𝑑𝑥

−1 3
1 9𝑥 2 6𝑥 3 𝑥4 1 9𝑥 2 6𝑥 3 𝑥4
= 16 ( + + ) + ( − + )
2 3 4 −3 16 2 3 4 1

1 9 81 81 1 81 84 9 1
= 16 [(2 − 2 + 1/4) − ( 2 − 54 + )] + 16 [( 2 − 54 + ) − (2 − 2 + 4)]
4 4
=0
Variance:
By def. 𝑉(𝑋) = 𝜎 2 = 𝐸(𝑋 2 ) − 𝜇 2

𝑉(𝑋) = ∫−∞ 𝑥 2 𝑓(𝑥)𝑑𝑥 − 𝜇 2

1 −1 1 1 1 3
= 16 ∫−3 (9𝑥 2 +𝑥 4 + 6𝑥 3 )𝑑𝑥 + 16 ∫−1(6𝑥 2 − 2𝑥 4 )𝑑𝑥 16 ∫1 (9𝑥 2 +𝑥 4 − 6𝑥 3 )𝑑𝑥

−1 1 3
1 9𝑥 3 6𝑥 4 𝑥5 2 6𝑥 3 𝑥5 1 9𝑥 3 6𝑥 4 𝑥5
= 16 ( + + ) + ( −2 5) + ( − + )
3 4 5 −3 16 3 0 16 3 4 5 1

= 14 − 0 = 14
4. A random variable X gives measurements of 𝐱 between 0 and 1 with a probability
𝟏𝟐𝐱 𝟑 − 𝟐𝟏𝐱 𝟐 + 𝟏𝟎𝐱, 𝐢𝐟 𝟎 ≤ 𝐱 ≤ 𝟏
density function 𝐟(𝐱) = {
𝟎, 𝐨𝐭𝐡𝐞𝐫𝐰𝐢𝐬𝐞
𝟏 𝟏
(i)Find 𝐏(𝐗 ≤ 𝟐) and P(X > ½ ) (ii) Find a number ′𝐤 ≤ 𝟏′ such that 𝐏(𝐗 ≤ 𝐤) = 𝟐:

1 1/2
Solution: P (X ≤ 2) = ∫−∞ f(x)dx
0 1/2
= ∫−∞ f(x)dx + ∫0 f(x)dx
1/2 1/2
= ∫−∞ (0 )dx + ∫0 (12x 3 − 21x 2 + 10x)dx
1/2
x4 x3 x2
= (12 − 21 + 10 2 )
4 3 0

1/2 3 7 5 9
= (3x 4 − 7x 3 + 5x 2 )0 =16 − 8 + 4 = 16

We know that P(−∞ < X < ∞) = 1


1 1 9 7
P (X > 2) = 1 − P (X ≤ 2) = 1 − 16 = 16
1
(ii) consider P(X ≤ k) = 2
k 1
⟹ ∫−∞ f(x)dx =
2
0 k 1
⟹ ∫−∞ f(x)dx + ∫0 f(x)dx = 2
0 k 1
⟹ ∫−∞(0)dx + ∫0 (12x 3 − 21x 2 + 10x)dx = 2
k
x4 x3 x2
⟹ (12 − 21 + 10 2 ) = 1/2
4 3 0

⟹ (3x 4 − 7x 3 + 5x 2 )k0 = 1/2


⟹ 3k 4 − 7k 3 + 5k 2 = 1/2
⟹ 6k 4 − 14k 3 + 10k 2 − 1 = 0
k is a real root of 6k 4 − 14k 3 + 10k 2 − 1 = 0 between 0 and 1
𝐤 = 𝟎. 𝟒𝟓𝟐

𝐤𝐱𝐞−𝛌𝐱 , 𝐱 ≥ 𝟎, 𝛌 > 𝟎
5. The probability density function 𝐟(𝐱) = {
𝟎, 𝐨𝐭𝐡𝐞𝐫𝐰𝐢𝐬𝐞
Determine (i)𝐤 (ii)mean (iii) Variance

Sol : We know that ∫−∞ f(x)dx = 1

⟹ ∫0 kxe−λx dx = 1

e−λx e−λx
⟹ k [x − ] =1
−λ λ2 0

1
⟹ k [0 − {0 − λ2}] = 1
k
⟹ λ2 = 1

⟹ k = λ2

(ii) Mean = ∫−∞ xf(x)dx

= ∫0 λ2 x 2 e−λx dx

e−λx e−λx e−λx
= λ2 [x 2 − 2x +2 ]
−λ λ2 −λ3 0

2 2
= λ2 [0 − {0 − λ3}] = λ

(iii) V(X) = σ2 = E(X 2 ) − μ2


∞ 2 2
V(X) = ∫−∞ 𝑥 2 𝑓(𝑥)𝑑𝑥 − (𝜆)

∞ 2 2
= ∫0 λ2 x 3 e−λx dx − (λ)

e−λx e−λx e−λx e−λx 2 2
= λ2 [x 3 − 3x 2 + 6x −6 ] − (λ)
−λ λ2 −λ3 λ4 0

6 6 2 2 2
= λ2 [0 − {0 − λ4}] = λ2 − (λ) = λ2
𝟎 , 𝐱 < 𝟐,
𝟐𝐱+𝟑
6. Is the function defined by 𝐟(𝐱) = { , 𝟐 ≤ 𝐱 ≤ 𝟒, a probability density function?
𝟏𝟖
𝟎 ,𝐱 > 𝟒
Find the probability that a variate X having 𝐟(𝐱) as density function will fall in the interval
𝟑 ≤ 𝐗 ≤ 4.

Sol: If f(x) is probability density function of a continuous random variable then (i) f(x) ≥ 0

(ii) ∫−∞ f(x)dx = 1

Clearly f(x) ≥ 0
∞ 2 4 ∞
∫−∞ f(x)dx = ∫−∞ f(x)dx + ∫2 f(x)dx + ∫4 f(x)dx
4
2x + 3
= 0+∫ dx + 0
2 18
1 2 1
= [x + 3x]4x=2 = [28 − 10] = 1
18 18
∴ f(x) is a probability density function.
4
(ii) P(3 ≤ X ≤ 4) = ∫3 f(x)dx
4 2x+3 1 1 5
= ∫3 dx =18 [x 2 + 3x]4x=3 = 18 [28 − 18] = 9
18

Probability that a variate X having f(x) as density function will fall in the interval 3 ≤ X ≤ 4
5
is .
9

7. The trouble shooting capacity of an IC chip in a circuit is a random variable X whose


𝟎 𝐟𝐨𝐫 𝐱 ≤ 𝟑
distribution function is given by 𝐅(𝐱) = { 𝟗 , where 𝐱 denotes the
𝟏 − 𝐱𝟐 𝐟𝐨𝐫 𝐱 > 𝟑
number of years.
Find the probability that the IC chip will work properly (i) Less than 8 years (ii) beyond 8
years (iii) between 5 to 9 yeas (iv) anywhere from 2 to 6 years.
x
Sol: We know that F(x) = P(X ≤ x) = ∫−∞ f(x)dx
9 55
(i) P(X ≤ 8) = F(8) = 1 − 82 = 64 = 0.8594

(ii) P(X > 8)= 1 − P(X ≤ 8)=1 − F(8) = 1 − 0.8594 = 0.1406


9 b
(iii) P(5 ≤ X ≤ 9) = ∫5 f(x)dx ∫a f(x)dx = F(b) − F(a)
9 5
= ∫−∞ f(x)dx − ∫−∞ f(x)dx
= F(9) − F(5)
9 9
= 1 − 92 − (1 − 52 )
9 9
= 25 − 81 = 0.2488

(iv) P(2 ≤ X ≤ 6) = F(6) − F(2)


9 3
= F(6) − 0 = 1 − 36 = 4 = 0.75

𝟎 𝐟𝐨𝐫 𝐱 ≤ 𝟎
8. If the pdf of a random variable X is given by 𝐟(𝐱) = { 𝟐 − 𝐱𝟑
.
𝟑𝐱 𝐞 𝐟𝐨𝐫 𝐱 > 𝟎
Determine the distribution function,
x
Sol: We know that F(x) = P(X ≤ x) = ∫−∞ f(x)dx

(i) For x ≤ 0
x x
F(x) = P(X ≤ x) = ∫−∞ f(x)dx = ∫−∞(0)dx = 0

(ii) For x > 0


x x 3
F(x) = P(X ≤ x) = ∫−∞ f(x)dx = ∫−∞ 3 x 2 e− x dx
x 3
= ∫0 3 x 2 e− x dx
x 3
= − ∫0 e− x d(− x 3 )
3 x 3
= −[e− x ]0 = 1 − e− x

𝟎 𝐟𝐨𝐫 𝐱 ≤ 𝟎
The Probability distribution function is 𝐅(𝐱) = { 𝟑
𝟏 − 𝐞− 𝐱 , 𝐟𝐨𝐫 𝐱 > 𝟎
9. The daily consumption of electric power is a random variable having probability density
𝟎 𝐟𝐨𝐫 𝐱 ≤ 𝟎
function 𝐟(𝐱) = {𝟏 −𝟐𝐱/𝟑 .
𝐱𝐞 𝐟𝐨𝐫 𝐱 > 𝟎
𝟗

If the total production is 12 million kw hours. Determine probability that there is a power
cut on any given day.
Sol: Let random variable X denotes consumption on a day,
If X takes value in between 0 and 12, then there will be no power cut.
If X takes value greater than 12, then there will be power cut.
We need to find P(X > 12)
12
P(−∞ ≤ X ≤ 12) = ∫−∞ f(x)dx
12 1
= ∫0 xe−2x/3 dx
9
12
1 e−2x/3 e−2x/3
= 9 [x − ]
−2/3 4/9 0

1 9 9
= 9 [−18e−8 − 4 e−8 + 4]=0.24925

P(X > 12) = 1 − P(−∞ ≤ X ≤ 12)=0.75075


The probability that there is a power cut on any given day is 0.75075
𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐞 𝐩𝐫𝐨𝐛𝐥𝐞𝐦𝐬
1 Suppose a continuous random variable X has the

probability density function

𝑓 (𝑥) = 𝐾 (1 − 𝑥 2 ), 𝑓𝑜𝑟 0 < 𝑥 < 3,

= 0 , 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

Find (i) K (ii) Mean (iii) Variance. (iv) Cumulative

distribution function (iv) 𝑃(1 < 𝑥 ≤ 2)

2 Suppose a continuous random variable X has the

probability density function

𝑓 (𝑥) = 𝐾 𝑥 2 𝑒 −𝑥 𝑤ℎ𝑒𝑛 𝑥 ≥ 0,

Find (i) K (ii) Mean (iii) Variance.

3 A continuous random variable X hat has density function

0, − ∞ < 𝑥 < 0,
𝑥2
𝑓(𝑥) = { , 0 ≤ 𝑥 < 3,
9
0, 𝑥 > 0.

(i)Determine the cumulative distribution function 𝐹(𝑥)

(ii) 𝑃(𝑋 < 0. 5) (ii) 𝑃(𝑋 > 1. 5) (iii) 𝑃(0.5 ≤ 𝑋 < 1. 5)

(iv) Mean

4. A continuous random variable X hat has distribution function

0, 𝑥 ≤ 1,
𝐹(𝑥) = {𝑘(𝑥 − 1)2 , 1 ≤ 𝑥 ≤ 3,
0, 𝑥 > 3.
Determine (i) pdf f(x) (ii)k (iii) Mean
5 The density function of an exponential random variable X is

𝜃𝑒 −𝜃𝑥 , 𝑥 ≥ 0
f (x) = 𝑓(𝑥) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

Find E (X), E (𝑋 2 ), Var (X).

6 The diameter of an electric cable, say X, is assumed to be a continuous


random variable with p.d.f :

f(x) = 6x (1-x), 0 ≤ x ≤ 1.

(i)Check that f(x) is p.d.f. and find mean, sd

(ii)Determine a number b such that P (X<b)= P (X>b).

7 Probability density function of a random variable X is


1
𝑠𝑖𝑛 𝑥 , 𝑓𝑜𝑟 0 ≤ 𝑥 ≤ 𝜋
𝑓(𝑥) = {2 .
0, 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

Find the mean and variance of the distribution.

Discrete Probability distributions


1. Discrete uniform distribution:
A discrete random variable 𝑋 has a uniform distribution if and only if its pmf is defined by
1
𝑝(𝑥) = 𝑛 for 𝑥 = 𝑥1 , 𝑥2 , … . . 𝑥𝑛 .

Then the random variable X is called as discrete uniform random variable. Its
distribution is as follows

𝑋=𝑥 𝑥1 𝑥2 ……………. 𝑥𝑛
𝑓(𝑥) 1 1 1
𝑛 𝑛 𝑛

BINOMIAL DISTRIBUTION

Binomial is a discrete probability distribution.

Suppose we perform a series of independent trials (that is, experiments


which can be repeated under identical circumstances) such that for each trial, 𝑝
is the probability of a success and 𝑞 that of a failure. Then, the probability of
getting exactly 𝑟 successes in 𝑛 trials is given by 𝑛𝐶𝑟 𝑝𝑟 𝑞 𝑛−𝑟 , where 𝑟 takes any
integral value form 0 to 𝑛. The probability of 0, 1, 2, …., 𝑟, …..,𝑛 success are
therefore given by 𝑞 𝑛 , 𝑛𝐶1 𝑝𝑞 𝑛−1 , 𝑛𝐶2 𝑝2 𝑞 𝑛−2 … . . , 𝑛𝐶𝑟 𝑝𝑟 𝑞 𝑛−𝑟 … , 𝑝𝑛 .

Definition : A random variable 𝑋 has a Binomial distribution if it assumes


only non-negative values and its probability mass function is given by

𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 ; 𝑥 = 0,1,2, … , 𝑛


𝑃(𝑋 = 𝑥) = 𝑝(𝑥) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

Where 𝑛= number of trials, p= probability of success, 𝑞 = 1 − 𝑝

𝑋= number of successes out of n trials.


The probability distribution of the number of successes 𝑃(𝑋 = 𝑥) so obtained is called the
Binomial Probability Distribution, because these probabilities are the successive terms in the
expansion of the binomial (𝑝 + 𝑞)𝑛 . 𝑋 is called as binomial variate and is denoted by
𝑋 ∼ 𝐵(𝑛, 𝑝), where 𝒏 and 𝑝 are called as parameters of the binomial
distribution. Sometimes, 𝑛 is also known as the degree of the distribution.
Condit ions of B inomial Distr ibution :

The Binomial D istr ibution holds under the following conditions:

1.Trials are re pe ate d under ide ntical c onditions for a finite positive inte ger, say 𝑛 times.

2.There are only two pos sible outcomes, e.g. succe ss or failure for e ach tr ial.

3.The pr obability of succe ss in each tr ial rem ains c onstant and does not c hange from tr ial to tr ial.

4.The tr ials are inde pe nde nt i. e., the probability of an eve nt in any trial is not affec ted by the res ults of
any other tr ial.

The constants of Binomial distribution

1. Mean 𝜇 = 𝑛𝑝 2. Variance 𝜎 2 = 𝑛𝑝𝑞 3. SD 𝜎 = √𝑛𝑝𝑞

Ex: A fair coin is tossed six times. Find the probability of getting four heads.

Sol: Number of trials 𝑛 = 6

𝑝= probability of getting head in a trial =1/2 , 𝑞 = 1 − 𝑝 = 1/2

1 4 1 2 6! 1 30 1
𝒑(𝟒) = 𝑷(𝑿 = 𝟒) =6𝐶4 𝑝4 𝑞6−4 = 6𝐶4 ( ) ( ) = = = 0.234
2 2 4!2! 26 2 64

Prob : Out of 800 families w ith 5 childre n e ach, how m any would you e xpec t to hav e 3 boys (b) 5 girls (c ) e ither 2
or 3 boy s (d) at leas t one boy. Assume equal probabilities for boy s and gir ls.

Solut ion:

Total num ber of families N = 80 0.

Number of c hildr en in eac h family n = 5

1
p = The probability of c hild to be a boy = P (m ale bir th) =
2

1
q = The probability of c hild to bea girl = P (fem ale bir th) =
2
(since equal probability for boys and girls )

𝑋: the num ber of boy s in eac h family.

The pr obability m ass function is

𝑝(𝑥) = 𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞𝑛−𝑥 , x =0,1,2,3,………, n

1 𝑥 1 5−𝑥
𝑝(𝑥) = 𝑃(𝑋 = 𝑥) = 5𝐶𝑥 ( ) ( ) , x=0,1,2,3,4,5
2 2

1
𝑝(𝑥) = 𝑃(𝑋 = 𝑥) = . 5𝐶𝑥 𝑥 = 0,1,2,3,4,5. 𝑝𝑒𝑟 𝑓𝑎𝑚𝑖𝑙𝑦
25

1 10 5
P(3 boys) = P (X=3) =p(3 ) = . 5𝐶3 = = 𝑝𝑒𝑟 𝑓𝑎𝑚𝑖𝑙𝑦
25 32 16

The expecte d of num ber of families hav ing 3 boys is

5
N. p(x) = (800) = 250 𝑓𝑎𝑚𝑖𝑙𝑖𝑒𝑠
16

1 1
P(5 gir ls) = P (no boys ) = P(X=0 ) =p(0 ) = . 5𝐶0 = 𝑝𝑒𝑟 𝑓𝑎𝑚𝑖𝑙𝑦
25 32

Thus for 800 fam ilies the probability of num ber of familie s having 5 gir ls

1
= (800) = 25 𝑓𝑎𝑚𝑖𝑙𝑖𝑒𝑠
32

P(either 2 or 3 boys ) = P(X = 2 ) +P(X = 3)

1 1
= p(2 ) +p(3) = . 5𝐶2 + . 5𝐶3
25 25

1 20 5
= (10 + 10) = = 𝑝𝑒𝑟 𝑓𝑎𝑚𝑖𝑙𝑦
25 32 8

5
∴ Expected num ber of fam ilies with 2 or 3 boys = (800) = 500 𝑓𝑎𝑚𝑖𝑙𝑖𝑒𝑠
8

P(at le as t one boy ) = p(X=1 )+P (X =2 ) +P (X=3) +P(X =4)+P(X=5 )

1
= (5 + 5𝐶2 + 5𝐶3 + 5𝐶4 + 5𝐶5 )
25 𝐶1

1
= (5 + 10 + 10 + 5 + 1)
25

31
=
32
31
∴ Expected num ber of fam ilies with at le ast one boy = (800) = 775
32

Prob 2 : Seven coins are tosse d and the num ber of he ads are note d. The expe rime nt is repe ate d 128 times and
the follow ing dis tr ibution is obtaine d.

No. of he ads 0 1 2 3 4 5 6 7 Total

Freque ncy 7 6 19 35 30 23 7 1 128

Fit a binom ial dis tr ibution ass uming.

(a) the coin is unbias ed. (b) the nature of the coin is not known.

Solut ion:
(a) the coin is unbiased.

1 1
P(X= occ urance of he ad), 𝑝 = ,𝑞 = 𝑎𝑛𝑑 𝑛 = 7
2 2

N= ∑ 𝑓𝑖 = 7 + 6 + 19 + 35 + 30 + 23 + 7 = 128

By binom ial dis tribution,

Pmf p(x) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞𝑛−𝑥 , 𝑥=0,1,2,….7

1 0 1 7 1
∴ 𝑝(0) = 7𝐶0 𝑝0 𝑞7 = 7𝐶0 ( ) ( ) =
2 2 27

PMF

at X =x.
No of he ads
Observed Expecte d or the oretical fre que ncy, E =
frequenc y P(x) N.P(x)
(X)

𝐸0 = 128.P(0)
1
0 7 p(0 ) =
27 1
= 128 𝑋 =1
27

𝐸1 = 128.P(1 )
7
1 6 p(1 ) =7.P (1) =
27 7
= 128 𝑋 =1
27

𝐸2 = 128.P(2)
21
2 10 p(2 ) =21.P (2) =
27 21
= 128 𝑋 =21
27

𝐸3 = 128.P(3)
35
3 35 p(3 ) =
27
=35

𝐸4 = 128.P(4)
35
4 30 p(4 ) =
27
=35

𝐸5 = 128.P(5)
21
5 23 p(5 ) =
27
=21

𝐸6 = 128.P(6
7
6 7 p(6 ) =
27
) =7

𝐸7 = 128.P(7)
1
7 1 p(7 ) =
27
=1

(b) T he nat ure of t he coin is not k now n.

We are give n n = 7, total frequency, N = ∑ 𝑓𝑖 = 128

∑ 𝑓𝑖 𝑋𝑖 6 + 38 + 105 + 120 + 115 + 42 + 7


∴ 𝑀𝑒𝑎𝑛 = =
∑ 𝑓𝑖 128

433
= = 3.383
128
i.e., mean E(X )=np = 3. 383 ⇒ 7𝑝 = 3.383

3.383
∴𝑝= = 0.4833 ≅ 0.48
7

And q = 1-p = 1-0.4833 = 0.51 6 7 ≅ 0.52

Hence the binom ial distr ibution to be fitted is given by the terms of

N(𝑞 + 𝑝)𝑛 = 128 (0.5167 + 0.4833)7

This information can als o be tabulate d as in the e arlier table.

POISSON DISTRIBUTION

Definition : A r andom var iable 𝑋 has a Poiss on distribution if it ass umes only non -ne gativ e v alues and its
probability m ass func tion is given by

𝜆𝑥
𝑒 −𝜆 ; 𝑥 = 0,1,2, … ,
𝑃(𝑋 = 𝑥) = 𝑝(𝑥) = { 𝑥!
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

It is denote d by X ~ 𝑃(𝜆), 𝜆 > 0 𝑖𝑠 𝑡ℎ𝑒 𝑝𝑎𝑟𝑎𝑚𝑒𝑡𝑒𝑟 of the distr ibution.

𝜆𝑥 𝜆𝑥
From normality, ∑∞ ∞
𝑥=0 𝑝(𝑥) = ∑𝑥=0 𝑒
−𝜆
= 𝑒 −𝜆 ∑∞
𝑥=0
𝑥! 𝑥!

𝜆1 𝜆2 𝜆3 𝜆4
= 𝑒 −𝜆 {1 + + + + + ⋯………….}
1! 2! 3! 4!

∑ 𝑝(𝑥) = 𝑒 −𝜆 . 𝑒 𝜆 = 1
𝑥=0

NOTE 1: A binom ial distr ibution te nds to a Poisson dis tr ibution under the follow ing conditions

1 n→ ∞ (i.e. the num ber of inde pendent tr ails are inde finite ly lar ge )

2 p→ 0 (i.e. the pr obability of succ ess for e ach tr ail is inde finite ly small)

𝜆
3 𝑛𝑝=𝜆, 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 ⟹ 𝑝=
𝑛

NOTE 2: For a Poisson dis tr ibution me an and v ar iance i.e. the me asure s of ce ntral tendency parame ter and the
measur es of disper sion param eter are c oincides w ith the par ameter 𝜆, 𝜇 = 𝐸(𝑋) = 𝑉(𝑋) = 𝜆

NOTE 3: T he Poiss on distr ibution ar ises w he n, in a continuous s pace or continuous tim e the occur rence of
events ar e discre te

APPLICATION :

Poiss on distr ibution is use d in quality contr ol statis tic s to c ount the number of defec ts of an item, I n ins ur ance
problems to c ount the num ber of c ausalitie s.

EX: It has been found that 2% of t he tools produced by a certain mac hine are defective. w hat is the probability
that in a shipment of 400 suc h tools 12 are defective.

Sol: Let x be number of defectives in 400 lot.

Range= {0,1,2,…………….., 400}

N=100, p =0.2 , q= 0.98

𝑃(𝑋 = 12) = 400𝐶12 (0.02)12 (0.98)388

Calculation is this because n is large and p is too small. So we use Poisson distribution.

λ = np = 400(0.02) = 8
λ12 812
𝑃(𝑋 = 12) = e−λ = e−8
12! 12!

Continuous Probability distributions


Continuous uniform distribution
Let X be a continuous random variable which takes values in between a and b .Then X is
said to follow uniform distribution if its probability function is given by
1
𝑝(x) = {𝑏 − 𝑎 for a ≤ x ≤ b
0, otherwise
Constants of the distribution:
∞ 𝑏 1
1. Mean = 𝐸(𝑋) = ∫−∞ 𝑥𝑓(𝑥)𝑑𝑥 = ∫𝑎 𝑥 𝑏−𝑎 𝑑𝑥
𝑏
1 𝑥2 1 𝑏 2 − 𝑎2 (𝑏 + 𝑎)
= [ ] = [ ]=
𝑏−𝑎 2 𝑎 𝑏−𝑎 2 2
∞ 𝑏 1
2. 𝐸(𝑋 2 ) = ∫−∞ 𝑥 2 𝑓(𝑥)𝑑𝑥 = ∫𝑎 𝑥 2 𝑏−𝑎 𝑑𝑥
𝑏
1 𝑥3 1 𝑏 3 − 𝑎3 (𝑏 2 + 𝑎𝑏 + 𝑎2 )
= [ ] = [ ]=
𝑏−𝑎 3 𝑎 𝑏−𝑎 3 3
2 (𝑏 2 +𝑎𝑏+𝑎2 ) (𝑏+𝑎)2 (𝑏−𝑎)2
Variance 𝑉(𝑋) = 𝐸(𝑋 2 ) − (𝐸(𝑋)) = − =
3 4 12

NORMAL DISTRIBUTION

Introduction:

In probability theory, a normal distribution is a continuous probability


distribution for a real-valued random variable.

Normal distributions are important in statistics and are often used in


the natural and social sciences to represent real-valued random variables whose
distributions are not known, due to the central limit theorem.
central lim it theorem: I t s tates that, under s ome c onditions , the aver age of m any sam ples (observ ations) of a
random v ar iable with finite me an and v ariance is itse lf a random v ariable, w hos e dis tribution c onver ges to a
norm al distr ibution as the num ber of sam ple s incre ases.

Therefore, phys ical quantities that are expec te d to be the s um of many indepe ndent proces ses, suc h
as me asur ement err ors, ofte n have distr ibutions that are nearly norm al.

Definition: A continuous random variable 𝑋 has a normal distribution if


its probability density function is given by
1 𝑥−𝜇 2
1
𝑒 − 2( )
𝑓 (𝑥) = 𝜎 , −∞ <𝑥 <∞
√2𝜋. 𝜎

It is denoted by 𝑋 ~ 𝑁(𝜇 , 𝜎 2 )

𝑊ℎ𝑒𝑟𝑒, 𝜇 = 𝑀𝑒𝑎𝑛 & 𝜎 2 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 are the two parameters.

A random variable which follows normal distribution is called as a normal variate and is
denoted by 𝑋 ~ 𝑁(𝜇 , 𝜎 2 ).
The curve 𝑦 = 𝑓(𝑥) which represents the normal distribution is called as normal curve.

If X ~ N(μ , σ2 ), then the probability for 𝑋 falls in (𝑎, 𝑏) is


1 x−μ 2
𝑏 1 𝑏
e− 2 ( )
𝑃(𝑎 < 𝑋 < 𝑏) = ∫𝑎 𝑓(𝑥)𝑑𝑥 = ∫ σ 𝑑𝑥
√2π. σ 𝑎

= Area under the normal curve 𝑦 = 𝑓(𝑥) between 𝑥 = 𝑎 and 𝑥 = 𝑏.



The total area under the normal curve 𝑦 = 𝑓(𝑥) is always 1. i.e ∫−∞ 𝑓(𝑥)𝑑𝑥 = 1

Constants of Normal distributions:


1. Mean =𝜇 2. Variance =σ2 3. Standard deviation =σ
It can be observed that Mean=Median=Mode in Normal distribution.
Note: Since it is a continuous distribution, the inclusion and exclusion of end points does not
impact the probability.
𝑃(𝑎 ≤ 𝑋 ≤ 𝑏) = 𝑃(𝑎 < 𝑋 ≤ 𝑏) = 𝑃(𝑎 ≤ 𝑋 < 𝑏) = 𝑃(𝑎 < 𝑋 < 𝑏)
Characteristics of the Normal distribution :

1. It is a continuous distribution.

Let X ~ N(μ , σ2 ). Then the pdf of X is


1 x−μ 2
1
e− 2( )
f (x) = σ , − ∞ < x < ∞,−∞ < μ < ∞, σ > 0
√2π. σ

2. The curve 𝑦 = 𝑓(𝑥) is called normal curve.

3. The curve is bell shaped symmetrical about the line 𝑥 = μ. The both tails of the curve
tend to infinity. The curve Symmetrical about the mean. Each half of the distribution
is a mirror image of the other half.

4. Area under the normal curve ∫−∞ 𝑓(𝑥)𝑑𝑥 = 1.

The area under the curve yields the probabilities, so the total of all
probabilities for a normal distribution is 1. Since the distribution is
symmetric, the area of the distribution on each side of the mean is 0.5.
5. Mean, Median and Mode of the distribution are coincide. So normal curve is
unimodal.
6. It is asymptotic to the horizontal axis. That is, it does not touch the x-axis
and it goes on forever in each direction.

7. The probability for X falls into the interval (a, b) is


1 x−μ 2
𝑏 1 − (𝑏 )
𝑃(𝑎 < 𝑋 < 𝑏) = ∫𝑎 𝑓(𝑥)𝑑𝑥 = ∫ e 2 σ 𝑑𝑥 (Area under the normal curve 𝑥 =
√2π. σ 𝑎
𝑎, 𝑥 = 𝑏)
STANDARD NORMAL DISTRIBUTION

A Normal distribution with mean, zero and standard deviation one is called as standard
normal distribution.
1 𝑥−𝜇 2
1
𝑒 −2 ( )
If 𝑋 ~ 𝑁(𝜇 , 𝜎 2 ) then pdf 𝑓 (𝑥) = 𝜎 , −∞ <𝑥 <∞
√2𝜋.𝜎

𝜇 = 𝐸(𝑋) & 𝜎 2 = 𝑉(𝑋)


𝑋−𝐸(𝑋) 𝑋−𝜇
Let us define a function 𝑍 = =
√𝑉(𝑋) 𝜎

𝑋−𝜇 1
Mean: 𝐸(𝑍) = 𝐸 ( ) = 𝜎 [𝐸(𝑋) − 𝐸(𝜇)]
𝜎
1
= 𝜎 [𝐸(𝑋) − 𝜇]

1
= 𝜎 (𝜇 − 𝜇) = 0

𝑋−𝜇 1 𝜇
Variance 𝑉(𝑍) = 𝑉 ( ) = 𝑉 [(𝜎) . 𝑋 − 𝜎]
𝜎

1 2
= (𝜎) 𝑉(𝑋) =1

𝑋−𝜇
Thus 𝒁 = ~ 𝑁(0 , 1).
𝜎

1
1 (z)2
The probability density function of 𝑧 is f (z) = e− 2 , −∞ <z<∞
√2π

The curve 𝑦 = 𝑓(𝑧) is called as standard normal curve.


𝑧
We denote the area ∫0 𝑘 𝑓(𝑧)𝑑𝑧 under the curve from 0 to 𝑧𝑘 by 𝐴(𝑧𝑘 )

Finding probability using area under standard Normal curve


1 x−μ 2
1
e− 2 ( )
Let X ~ N(μ , σ2 ) then f (x) = σ
√2π.σ

Model-1 𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥2 )
x1 −μ x2 −μ
𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥2 ) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 ) where 𝑧1 = , 𝑧2 =
σ σ

Case 1. If 𝑧1 , 𝑧2 are posiive.


𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥2 ) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )
𝑧
= ∫𝑧 2 𝑓(𝑧)𝑑𝑧
1

𝑧2 𝑧
= ∫0 𝑓(𝑧)𝑑𝑧 − ∫0 1 𝑓(𝑧)𝑑𝑧

= 𝐴(𝑧2 ) − 𝐴(𝑧1 )
Case 2. If 𝑧1 < 0 𝑎𝑛𝑑 𝑧2 > 0.
𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥2 ) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )
𝑧
= ∫𝑧 2 𝑓(𝑧)𝑑𝑧
1

0 𝑧 −𝑧1 𝑧
= ∫𝑧 𝑓(𝑧)𝑑𝑧 + ∫0 2 𝑓(𝑧)𝑑𝑧 = ∫0 𝑓(𝑧)𝑑𝑧 + ∫0 2 𝑓(𝑧)𝑑𝑧
1

= 𝐴(−𝑧1 ) + 𝐴(𝑧2 )
Case 3. If 𝑧1 < 0 𝑎𝑛𝑑 𝑧2 < 0.
𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥2 ) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )
𝑧
= ∫𝑧 2 𝑓(𝑧)𝑑𝑧
1

0 0 −𝑧1 −𝑧2
= ∫𝑧 𝑓(𝑧)𝑑𝑧 − ∫𝑧 𝑓(𝑧)𝑑𝑧 = ∫0 𝑓(𝑧)𝑑𝑧 − ∫0 𝑓(𝑧)𝑑𝑧
1 2

= 𝐴(−𝑧1 ) − 𝐴(−𝑧2 )
Model-2 𝑃(𝑥1 ≤ 𝑋)
x1 −μ
𝑃(𝑥1 ≤ 𝑋) = 𝑃(𝑧1 ≤ 𝑍) where 𝑧1 = ,
σ
Case 1. If 𝑧1 > 0.
𝑃(𝑥1 ≤ 𝑋) = 𝑃(𝑧1 ≤ 𝑍)
∞ 1 𝑧
= ∫𝑧 𝑓(𝑧)𝑑𝑧 = 2 − ∫0 1 𝑓(𝑧)𝑑𝑧
1

1
= 2 − 𝐴(𝑧1 )

Case 2. If 𝑧1 < 0.
𝑃(𝑥1 ≤ 𝑋) = 𝑃(𝑧1 ≤ 𝑍)

= ∫𝑧 𝑓(𝑧)𝑑𝑧
1 𝑧1
0 1 −𝑧1 1
= ∫𝑧1 𝑓(𝑧)𝑑𝑧 +2= ∫0 𝑓(𝑧)𝑑𝑧 +2
1
= + 𝐴(−𝑧1 )
2

Model-3 𝑃(𝑋 ≤ 𝑥2 )
x2 −μ
𝑃(𝑋 ≤ 𝑥2 ) = 𝑃(𝑍 ≤ 𝑧2 ) where 𝑧2 = ,
σ

Case 1. If 𝑧2 > 0.
2 𝑧
𝑃(𝑋 ≤ 𝑥2 ) = 𝑃(𝑍 ≤ 𝑧2 ) = ∫−∞ 𝑓(𝑧)𝑑𝑧
1 𝑧 1
= 2 + ∫0 2 𝑓(𝑧)𝑑𝑧 = 2 + 𝐴(𝑧2 )

Case 2. If 𝑧2 < 0.
2 𝑧
𝑃(𝑋 ≤ 𝑥2 ) = 𝑃(𝑍 ≤ 𝑧2 ) = ∫−∞ 𝑓(𝑧)𝑑𝑧
1 2
= 2 − ∫𝑧 𝑓(𝑧)𝑑𝑧
2

1 0 1 −𝑧2
= 2 − ∫𝑧 𝑓(𝑧)𝑑𝑧 = 2 − ∫0 𝑓(𝑧)𝑑𝑧
2

1
= 2 − 𝐴(−𝑧2 )

Problems on Normal distributions:


Prob 1: If X is a normal variate, find the area A (i) to the left of 𝒛 = −𝟏. 𝟕𝟖 (ii) to the right
of 𝒛 = −𝟏. 𝟒𝟓 (iii)corresponding to −𝟎. 𝟖 ≤ 𝒛 ≤ 𝟏. 𝟓𝟑 (iv) to the left of 𝒛 = −𝟐. 𝟓𝟏
and to the right of 𝒛 = 𝟏. 𝟖𝟑
Ans: (i) 𝐴 = 𝑃(𝑍 < −1.78)
1
= 2 − 𝑃(−1.78 < 𝑍 < 0)
1
= 2 − 𝑃(0 < 𝑍 < 1.78)

1 1
= 2 − 𝑃(0 < 𝑍 < 1.78) = 2 − 𝐴(1.78) = 0.5 − 0.4625 = 0.0375

(ii)𝐴 = 𝑃(𝑍 > −1.45)


1
= 2 + 𝑃(−1.45 < 𝑍 < 0)
1
= 2 + 𝑃(0 < 𝑍 < 1.45)
1
= 2 + 𝐴(1.45) = 0.5 + 0.4265 = 0.9265

(iii)) 𝐴 = 𝑃(−0.8 < 𝑍 < 1.53)


= 𝑃(−0.8 < 𝑍 < 0) + 𝑃(0 < 𝑍 < 1.53)
= 𝑃(0 < 𝑍 < 0.8) + 𝑃(0 < 𝑍 < 1.53)
= 𝐴(0.8) + 𝐴(1.53)
= 0.2881 + 0.437 = .7251
(iV)𝐴 = 𝑃(−2.52 < 𝑍) + 𝑃(𝑍 > 1.83)
1 1
= 2 − 𝑃(−0.252 < 𝑍 < 0) + 2 − 𝑃(0 < 𝑍 < 1.83)

= 0.5 − 𝑃(0 < 𝑍 < 2.52) + 0.5 − 𝑃(0 < 𝑍 < 1.83)
= 1 − 𝐴(0.252) − 𝐴(1.83)
= 1 − 0.4941 − 0.4664 = 0.0395

Prob 2. If 𝑿 is a normal variate with mean 30 and standard deviation 5. Find the
probabilities that 𝟐𝟔 ≤ 𝑿 ≤ 𝟒𝟎 (ii)𝑿 ≥ 𝟒𝟓
Sol: Given that mean 𝜇 = 30, standard deviation 𝜎 = 5
𝑋−𝜇
Standard normal variate 𝑧 =
𝜎

26−𝜇 26−30 40−𝜇 40−30


(i) 𝑧1 = = = −0.8, 𝑧2 = = =2
𝜎 5 𝜎 5

𝑃(26 ≤ 𝑋 ≤ 40) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )

= 𝑃(−0.8 ≤ 𝑍 ≤ 2) 2

= 𝑃(−0.8 ≤ 𝑍 ≤ 0) + 𝑃(0 ≤ 𝑍 ≤ 2)
= 𝑃(0 ≤ 𝑍 ≤ 0.8) + 𝑃(0 ≤ 𝑍 ≤ 2)

=A(0.8)+A(2)

= 0.2881+0.4772=0.7653

45−𝜇 45−30
(ii) 𝑧1 = = =3
𝜎 5

𝑃(45 ≤ 𝑋) = 𝑃(𝑧1 ≤ 𝑍)

= 𝑃(3 ≤ 𝑍) = 𝑃(𝑍 ≥ 3) 3

= ½ −𝑃(0 ≤ 𝑍 ≤ 3) = ½ - A(3)

= 0.5-0.4987=0.0013

Prob 3: The mean and standard deviation of the marks obtained by 1000 students in a
examination are respectively 34.5 and 16.5 . Assuming the normality of the distribution ,
find the approximate number of students expected to obtain marks between 30 and 60.
Sol: Let 𝑋 be the random variable which denotes the marks obtained by a student.
Given that mean 𝜇 = 34.5, standard deviation 𝜎 = 16.5
𝑋−𝜇
Standard normal variate 𝑧 = 𝜎

To find the probability of student obtains marks in between 30 and 60


i.e 𝑃(30 < 𝑋 < 60)
30−𝜇 30−34.5 60−𝜇 60−34.5
𝑧1 = = = −0.27, 𝑧2 = = = 1.54
𝜎 16.5 𝜎 16.5

𝑃(30 ≤ 𝑋 ≤ 60) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )

= 𝑃(−0.27 ≤ 𝑍 ≤ 1.54)

= 𝑃(−0.27 ≤ 𝑍 ≤ 0) + 𝑃(0 ≤ 𝑍 ≤ 1.54)

= 𝑃(0 ≤ 𝑍 ≤ 0.27) + 𝑃(0 ≤ 𝑍 ≤ 1.54) -0.27 1.54

=A(0.27)+A(1.54)

= 0.1064+0.4382=0.5446

=A(0.27)+A(1.54)
= 0.1064+04382=0.5446

𝑓𝑎𝑣𝑜𝑢𝑟𝑎𝑏𝑙𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠 𝑜𝑏𝑡𝑎𝑖𝑛 𝑚𝑎𝑟𝑘𝑠 𝑖𝑛 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 30&60


𝑃(30 ≤ 𝑋 ≤ 60) = 𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠
𝑛
0.5446 = 1000

Expected number of students obtains marks between 30 and 60 is n = 544.6~ 545

Prob 4: A manufacturer knows from experience that the resistance of resistors he


produces is follows normal distribution with mean 100 ohms and standard deviation 2
ohms. What percentage of resistors will have resistance between 98 ohms and 102 ohms?
Sol: Let 𝑋 be the random variable which denotes the resistance produced by a resistor.
Given that mean 𝜇 = 100, standard deviation 𝜎 = 2
𝑋−𝜇
Standard normal variate 𝑧 = 𝜎

To find probability 𝑃(98 < 𝑋 < 102), probability of a resistor with resistance in
between 98 and 102

98−𝜇 98−100 102−𝜇 102−100


𝑧1 = = = −1, 𝑧2 = = =1
𝜎 2 𝜎 2

𝑃(98 < 𝑋 < 102) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )

= 𝑃(−1 ≤ 𝑍 ≤ 1) -1 1

= 𝑃(−1 ≤ 𝑍 ≤ 0) + 𝑃(0 ≤ 𝑍 ≤ 1)

= 𝑃(0 ≤ 𝑍 ≤ 1) + 𝑃(0 ≤ 𝑍 ≤ 1)

=A(1) + A(1)

= 0.3413+0.3413=0.6826

The percentage of resistors having resistance between 98 and 100 is


𝑃(98 < 𝑋 < 102) × 100 = 68.26%

Prob 5. A sales tax officer has reported that the average sales of the 500 business that he
has to deal with during a year is 36,000 Rs with standard deviation of 10,000 Rs . Assuming
that the sales in these business are normally distributed, find ( i) the number of business
as the sales of which are 40,000 Rs and above (ii) the percentage of business the sales of
which are likely to range between 30,000 Rs and 40,000 Rs.
Sol: Let X be the sales in the year by a business firm.
Given that mean 𝜇 = 36,000, standard deviation 𝜎 = 10,000
𝑋−𝜇
Standard normal variate 𝑧 = 𝜎

(i) To find probability 𝑃(𝑋 ≥ 40,000),

40,000−𝜇 40,000−36,000
z= = = 0.4,
𝜎 10,000

𝑃(𝑋 ≥ 40,000) = 𝑃(𝑍 ≥ 𝑧)

= 𝑃(𝑍 ≥ 0.4)
0.4
=0.5 - 𝑃(0 ≤ 𝑍 ≤ 0.4)

= 0.5 - 0.1554 = 0.3446


Number of business firms with sales 40,000 and above is
𝑃(𝑋 ≥ 40,000) × 500 = 172.3 ≅ 172

(ii) To find the probability of a business with sales in between 30,000 and 40,000
i.e 𝑃(30,000 < 𝑋 < 40,000).
30,000−𝜇 30,000−36,000 40,000−𝜇 40,000−36,000
𝑧1 = = = −0.6, 𝑧2 = = = 0.4
𝜎 10,000 𝜎 10,000

𝑃(30,000 < 𝑋 < 40,000) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )

= 𝑃(−0.6 ≤ 𝑍 ≤ 0.4)

= 𝑃(−0.6 ≤ 𝑍 ≤ 0) + 𝑃(0 ≤ 𝑍 ≤ 0.4) -0.6 0.4

= 𝐴(0.6) + 𝐴(0.4)

= 0.2258 + 0.1554 = 0.38816826

The percentage of business firms having sales between 30,000 and 40,000
is 𝑃(30,000 < 𝑋 < 40,000) × 100 = 38.11%

Prob 6: In normal distribution , 7% of the items are under 35 and 89% of the items are
under 63. Determine mean and variance.
Sol: From the given distribution , we can draw the distribution of the total items under the
normal curve as follows. We know that the normal curve is symmetric about the mean.

Given 7% under 35 , so 43% are between 35 and 𝜇.


Given 89% under 63 , so 11% above 63.
𝑋−𝜇
Let 𝑧 = , Then we get standard normal distribution of the items as follows
𝜎

0.43 0.39
0.11
0.07

z1 z2

Let 𝑥1 = 35 and 𝑥2 = 63 and


0.43 = 𝑃(𝑧1 ≤ 𝑍 ≤ 0) = 𝑃(0 ≤ 𝑍 ≤ −𝑧1 ) = 𝐴(−𝑧1 )
𝐴(−𝑧1 ) = 04306
⟹ −𝑧1 = 1.48 ⟹ 𝑧1 = −1.48
0.39 = 𝑃(0 ≤ 𝑍 ≤ 𝑧2 ) = 𝐴(𝑧2 )
𝐴(𝑧2 ) = 0.39 = 0.3907
⟹ 𝑧2 = 1.23
𝑥1 −𝜇 𝑥2 −𝜇
= −1.48 and = 1.23
𝜎 𝜎
35−𝜇 63−𝜇
i.e = −1.48 and = 1.23
𝜎 𝜎

⟹ 35 − 𝜇 = −1.48𝜎 𝑎𝑛𝑑 63 − 𝜇 = 1.23𝜎


⟹ 2.71𝜎 = 28 ⟹ 𝜎 = 10.332
𝜇 = 35 + 1.48𝜎 = 35 + 1.48 × 10.332 = 50.3
Hence Mean 𝜇 = 50.3 and Variance𝜎 2 = 10.3322 = 106.75
Prob 7. In normal distribution , 31% of the items are under 45 and 8% of the items are
over 64. Find mean and variance of the distribution.
Sol: From the given distribution, we can draw the distribution of the total items under the
normal curve as follows. We know that the normal curve is symmetric about the mean 𝜇.

Given 31% under 45 , so 19 % are between 45 and 𝜇.


Given 8% above 64 , so 42% between 𝜇 and 6.
𝑋−𝜇
Let 𝑧 = , then we get standard normal distribution of the items as follows
𝜎

0.19 0.42

0.31 0.08

z1 z2

Let 𝑥1 = 45 and 𝑥2 = 64
𝑃(𝑧1 ≤ 𝑍 ≤ 0) = 0.19 = 0.1915
𝑃(𝑧1 ≤ 𝑍 ≤ 0) = 𝑃(0 ≤ 𝑍 ≤ −𝑧1 ) = 𝐴(−𝑧1 ) = 0.1915
⟹ −𝑧1 = 0.5 ⟹ 𝑧1 = −0.5
𝑃(0 ≤ 𝑍 ≤ 𝑧2 ) = 𝐴(𝑧2 ) = 0.42 = 0.4207
⟹ 𝑧2 = 1.41
𝑥1 −𝜇 𝑥2 −𝜇
= −0.5and = 1.41
𝜎 𝜎
45−𝜇 64−𝜇
i.e = −0.5 and =1.41
𝜎 𝜎

⟹ 45 − 𝜇 = −0.5𝜎 𝑎𝑛𝑑 64 − 𝜇 = 1.41𝜎


⟹ 1.91𝜎 = 19 ⟹ 𝜎 = 9.94
𝜇 = 45 + 0.5𝜎 = 45 + 0.5 × 9.94 = 49.97
Hence Mean 𝜇 = 49.97 and Variance𝜎 2 = 9.942 = 98.8
Prob 8: The marks obtained in mathematics by 1000 students is normally distributed with
mean 78% and standard deviation 11%. Determine
(i) How many students got marks above 90%
(ii) What was the highest marks obtained by the lowest 10% of the students.
(iii) Within what limits did the middle of 90% of the students lie.
Sol: Let X be the random variable which denotes percentage of marks obtained by a
student .
Given that mean 𝜇 = 0.78, standard deviation 𝜎 = 0.11
𝑋−𝜇
Standard normal variate 𝑧 =
𝜎

(i) To find probability 𝑃(𝑋 ≥ 90), probability that a student at random got above 90%
90−𝜇 90−78
z= = = 1.09
𝜎 11

𝑃(𝑋 ≥ 90) = 𝑃(𝑍 ≥ 1.09)

= 0.5 − 𝑃(0 ≤ 𝑍 ≤ 1.09) 1.09

= 0.5 − 𝐴(1.09) = 0.5 − 0.3621 = 0.1379

Number of students who got above 90% = 𝑃(𝑋 ≥ 90) × 1000 = 137.9 ≅ 138

(ii) Here we want to find highest percentage of marks 𝑥1 , by the lowest 10% of the
students.

𝑋−𝜇 𝑥1 −𝜇
𝑍= , 𝑧1 =
𝜎 𝜎

𝑃(𝑋 < 𝑥1 ) = 𝑃(𝑍 < 𝑧1 ) = 0.1

Clearly 𝑧1 < 0

0.1
𝑃(𝑧1 < 𝑍 < 0) = 0.5 − 0.1 = 0.4

z1
𝑃(0 < 𝑍 < −𝑧1 ) = 𝑃(𝑧1 < 𝑍 < 0) = 0.4 = 0.4015

i.e 𝐴(−𝑧1 ) = 0.4015

⟹ −𝑧1 = 1.29 ⟹ 𝑧1 = −1.29

𝑥1 −𝜇
We have 𝑧1 = 𝜎

𝑥1 −0.78
⟹ −1.29 = 0.11

⟹ 𝑥1 = 0.78 − 1.29 × 0.11 = 0.6392

Thus highest marks obtained by the lowest 10% of students is 0.6392 × 100 = 63.92 ≅ 64.

(iii) Let the middle 90% of students obtained marks between 𝑥1 𝑎𝑛𝑑 𝑥2 .
𝑥1 −𝜇 𝑥2 −𝜇
𝑧1 = , 𝑧2 =
𝜎 𝜎

𝑃(𝑥1 < 𝑋 < 𝑥2 ) = 𝑃(𝑧1 < 𝑍 < 𝑧2 ) = 0.9


𝑧1 < 0, 𝑧2 > 0
0.45
0.4
𝑃(𝑧1 < 𝑍) + 𝑃(𝑍 < 𝑧2 ) = 0.1
0.05 0.05

Since we consider middle 90%, z1 z2


0.1
𝑃(𝑍 > 𝑧1 ) = 𝑃(𝑍 > 𝑧2 ) = = 0.05
2

𝑃(𝑍 < 𝑧1 ) = 𝑃(𝑍 > −𝑧1 ) = 0.05 ⟹ 𝑃(0 < 𝑍 < −𝑧1 ) = 0.45 = 0.4505
𝐴(−𝑧1 ) = 0.4505 ⟹ −𝑧1 = 1.64 ⟹ 𝑧1 = −1.64
𝑥1 −0.78
⟹ = −1.64 ⟹ 𝑥1 = 0.78 − 0.11 × 1.64 = 0.5996
0.11

𝑃(𝑍 > 𝑧2 ) = 0.05 ⟹ 𝑃(0 < 𝑍 < 𝑧2 ) = 0.45


𝑥2 −0.78
⟹ = 1.64
0.11

⟹ 𝑥2 = 0.78 + 0.11 × 1.64 = 0.9604


Hence the middle 90% of students have marks between 59.96 and 96.04
i.e approximately between 60 to 96 marks.
CHARACTERISTICS of Normal D istribut ion:

MOMENT GENERATING FUNCTION

1 𝑥−𝜇 2
1 ( )
If 𝑋 ~ 𝑁(𝜇 , 𝜎 2 ), pdf is 𝑓 (𝑥) = 𝑒 −2 𝜎 , −∞ < 𝑥 <∞
√2𝜋.𝜎

𝒕 𝟐 𝝈𝟐
𝑴𝑮𝑭, 𝑴𝑿 (𝒕) = 𝒆𝒕𝝁+ 𝟐

Moments (me an and variance)

The Re lationship be tween mgf and mome nts,

𝑑𝑟
𝐸 (𝑋 𝑟 ) = [𝑀𝑋 (𝑡)]𝑡=0 , 𝑟 = 1,2, … . . ,.
𝑑𝑡 𝑟

𝑑 𝒕 𝟐 𝝈𝟐
Mean ( 𝝁):𝐸 (𝑋) = { 𝒆𝒕𝝁+ 𝟐 }
𝑑𝑡
𝑡=0

𝒕 𝟐 𝝈𝟐 𝑑 𝒕𝟐 𝝈𝟐
= 𝒆𝒕𝝁+ 𝟐 . [𝒕𝝁 + ] , 𝑎𝑡 𝑡 = 0
𝑑𝑡 𝟐

𝒕 𝟐 𝝈𝟐
𝟐𝒕 𝝈𝟐
= 𝒆𝒕𝝁+ 𝟐 . { 𝝁. 𝟏 + }, at t=0
𝟐

𝟎 𝟐 𝝈𝟐 𝟐.𝟎 𝝈𝟐
= 𝒆𝟎 𝝁+ 𝟐 . {𝝁 + } = 𝒆𝟎 . { 𝝁}
𝟐

E(X) = 𝝁

B) Variance :

𝜎 2 = 𝑉(𝑋) = 𝐸(𝑋 2 ) − 𝑚𝑒𝑎𝑛2

𝑑2
𝐸(𝑋 2 ) = [𝑀𝑋 (𝑡)]𝑡=0
𝑑𝑡 2

𝑑 𝑑 𝑡2 𝜎 2
= { ( 𝑒 𝑡𝜇+ 2 )}
𝑑𝑡 𝑑𝑡
𝑡=0

𝒕𝟐 𝝈𝟐 𝒕 𝟐 𝝈𝟐
={ 𝒆𝒕𝝁+ 𝟐 { 𝝈𝟐 } + 𝒆𝒕𝝁+ 𝟐 {𝝁. 𝟏 + 𝒕 𝝈𝟐 }{𝝁. 𝟏 + 𝒕 𝝈𝟐 } }𝒕=𝟎

= 𝝈𝟐 + 𝝁𝟐

V(X) = E(X 2 ) − E(X)2 = 𝝈𝟐 + 𝝁𝟐 − 𝝁𝟐 = 𝝈𝟐


1 𝑥−𝜇 2
1 ( )
If 𝑋 ~ 𝑁(𝜇 , 𝜎 2 ), pdf, 𝑓 (𝑥) = 𝑒 −2 𝜎 , −∞ < 𝑥 <∞
√2𝜋.𝜎

𝐸(𝑋) = 𝜇 & 𝑉(𝑋) = 𝜎 2

Normal Approximation to Binomial distribution:

Let 𝑋~𝐵(𝑛, 𝑝). Since X is Binomial variate, its probability mass function is

𝑝(𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞𝑛−𝑥 ; 𝑥 = 0,1,2, … , 𝑛.

Then 𝑃(𝑥1 ≤ 𝑋 ≤ 𝑥𝑛 ) = 𝑝(𝑥1 ) + 𝑝(𝑥2 ) + ⋯ . . 𝑝(𝑥𝑛 )

= 𝑛𝐶𝑥1 𝑝 𝑥1 𝑞 𝑛−𝑥1 + 𝑛𝐶𝑥2 𝑝 𝑥2 𝑞𝑛−𝑥2 + ⋯ … . +𝑛𝐶𝑥𝑛 𝑝 𝑥𝑛 𝑞𝑛−𝑥𝑛 .

For large values of n, the calculation of above binomial probabilities are difficult. In such cases the binomial curve is approximated by a
normal curve and the required probability is computed as follows.

Here 𝑋 ≅ 𝑍 [ Where 𝑋 is binomial variate and Z is standard normal variate]

𝑃(𝑥1 < 𝑋 < 𝑥2 ) = 𝑃(𝑧1 < 𝑍 < 𝑧2 ).


(𝑥1 −1/2)−𝜇 (𝑥1 −1/2)−𝑛𝑝 (𝑥2 +1/2)−𝜇 (𝑥2 +1/2)−𝑛𝑝
Where 𝑧1 = = , 𝑧2 = =
𝜎 √𝑛𝑝𝑞 𝜎 √𝑛𝑝𝑞

𝑧2
𝑃(𝑥1 < 𝑋 < 𝑥2 ) = ∫𝑧1 𝑓(𝑧)𝑑𝑧 = area under the normal curve 𝑦 = 𝑓(𝑧), between 𝑧1 ,𝑧2 .

Prob1. Find the probability by guess work a student can correctly answer 25 to 30 questions in a multiple choice quiz consisting of 80
questions. Assume that in each question with four choices, only one choice is correct and student has no knowledge of the subject.
1 3
Sol: Given that 𝑝 = , 𝑞 = , 𝑛 = 80.
4 4

Let 𝑋= number of correct answers out of 80.

Range of 𝑋 = {0, 1, 2 … … 80}


1
Clearly 𝑋~𝐵(80, )
4

∴ 𝑃(25 ≤ 𝑋 ≤ 30) = 𝑃(25) + 𝑃(26) + ⋯ … . . +𝑃(30)

1 25 3 55 1 26 3 54 1 30 3 50
= 80𝐶25 ( ) ( ) + 80𝐶26 ( ) ( ) + ⋯ . . +80𝐶30 ( ) ( )
4 4 4 4 4 4

Clearly Binomial probabilities are difficult to evaluate.

Let 𝑋 ≅ 𝑍 [ Where Z is standard normal variate]

𝑃(25 ≤ 𝑋 ≤ 30) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )


(25−1/2)−𝜇 24.5−𝑛𝑝 24.5−20 4.5 (30+1/2)−𝜇 30.5−𝑛𝑝 30.5−20 10.5
𝑧1 = = = = = 1.16., 𝑧2 = = = = = 2.71
𝜎 √𝑛𝑝𝑞 √15 3.87 𝜎 √𝑛𝑝𝑞 √15 3.87

𝑃(25 ≤ 𝑋 ≤ 30) = 𝑃(1.16 ≤ 𝑍 ≤ 2.71)


𝟐.𝟕𝟏
= ∫𝟏.𝟏𝟔 𝒇(𝒛)𝒅𝒛 = 𝑨(𝟐. 𝟕𝟏) − 𝑨(𝟏. 𝟏𝟔) = 𝟎. 𝟒𝟗𝟔𝟔 − 𝟎. 𝟑𝟕𝟕 = 𝟎. 𝟏𝟏𝟗𝟔

Prob 2. Find the probability that out of 100 patients between 84 and 95 inclusive will survive

a heart-operation, given that chances of survival is 0.9.

Sol: Given that probability of survival is 𝑝 = 0.9.

𝑞 = 1 − 𝑝 = 0.1, 𝑛 = 100

Let 𝑋= number of survivals out of 100 patients.

Range of 𝑋 = {0, 1, 2 … … 100}

Clearly 𝑋~𝐵(100, 0.9)

∴ 𝑃(84 ≤ 𝑋 ≤ 95) = 𝑃(84) + 𝑃(85) + ⋯ … . . +𝑃(95)

= 100𝐶84 (0.9)84 (0.1)16 + 100𝐶85 (0.9)85 (0.1)15 + ⋯ . . +100𝐶95 (0.9)95 (0.1)5

Evaluation in this way is difficult

Normal approximation:

Let 𝑋 ≅ 𝑍 [ Where Z is standard normal variate]

𝑃(84 ≤ 𝑋 ≤ 95) = 𝑃(𝑧1 ≤ 𝑍 ≤ 𝑧2 )


(84−1/2)−𝜇 83.5−𝑛𝑝 83.5−90 −6.5
𝑧1 = = = = = −2.17,
𝜎 √𝑛𝑝𝑞 √9 3

(95+1/2)−𝜇 95.5−𝑛𝑝 95.5−90 5.5


𝑧1 = = = = = 1.83
𝜎 √𝑛𝑝𝑞 √9 3

𝑃(84 ≤ 𝑋 ≤ 95) = 𝑃(−2.17 ≤ 𝑍 ≤ 1.83)


1.83 0 1.83 2.17 1.83
= ∫−2.17 𝑓(𝑧)𝑑𝑧 = ∫−2.17 𝑓(𝑧)𝑑𝑧 + ∫0 𝑓(𝑧)𝑑𝑧 = ∫0 𝑓(𝑧)𝑑𝑧 + ∫0 𝑓(𝑧)𝑑𝑧

= 𝐴(2.17) + 𝐴(1.83) = 0.485 + 0.4664 = 0.9514


Practice problems:

1. If X is uniformly distributed random variable that takes values between 0 and 2 . Then find
𝐸(𝑋 4 ) .

2. If the salaries of 400 employees in a company are normally distributed with mean 65,000
and standard deviation 10,000. How many employees get salary (i) greater than 72,000 (ii)
less than or equal to 64,00 (iii)between 60 and 70 thousands

3. If 𝑋 is a normal variate with mean 35 and SD 5, then find (I)𝑃(25 < 𝑋 < 40) (ii) 𝑃(𝑋 ≥ 44)

If X is a random variable following normal distribution with mean 1 and variance 4. Let Y be another
normal variate with men -1 and variance unknown. If 𝑃(𝑋 ≤ −1) = 𝑃(𝑌 ≥≥ 2), then find the
variance of 𝑌.

4. The wages of 20,000 employees was found normally distributed withs mean 7500 and
standard deviation 500. Find (i) the percentage of employee with salary exceeding 6680
(ii)percentage of employees with salary exceeding 8320(iii) number of employees with salary
between 6800 and 7900.

5. Suppose that the weights of 800 male students are normally distributed with mean 42
pounds and standard deviation 10 pounds. Find the number of students whose weights 52
pounds.

6. Eight coins are tossed together. Find by normal approximation the probability of getting 1 to
4 heads in a single toss

7. Determine the probability that getting an even number on the face 3 to 5 times in throwing
10 dice together by normal approximation

You might also like