Discrete random variable

A discrete random variable is a random variable that can assume only distinct values.

If we toss a coin twice, the number of heads obtained could be 0,1or 2.The probabilities
of these occurring are as follows.

P (no heads) = P (TT) = (0.5)(0.5) = 0.25

P (one head) = P (HT) + P (TH) = (0.5) (0.5) + (0.5) (0.5) = 0.5
P (two heads) = P (HH) = (0.5) (0.5) =0.25

We can show the results on a table, known as probability distribution.

Number of heads 0 1 2
Probability 0.25 0.5 0.25

The variable being considered is the ‘ the number of heads obtained in two tosses’ and it
can be denoted by X. It can only take exact values 0,1 or 2 and so is called a discrete
variable.

The probabilities can be written P (X=0) = 0.25, P (X=1) = 0.5, P (X=2) =0.25
Now if the sum of the probabilities is 1, the variable is said to be random

In this example
P (X=0) + P (X=1) + P (X=2) = 1
So X is a random variable.
The probability distribution is often written

X 0 1 2
P(X=x 0.25 0.5 0.25
)

And the statement ‘ the sum of the probabilities is 1’ is written

 P( X
allx
 x)  1

We usually denote a random variable (r .v) by a capital letter ( X , Y, Z etc. ) and the
particular value it takes by a small letter ( x, y, r etc.).

Example

An executive recruiter is interviewing three candidates. Let X be the discrete random

variable that denotes the number of males interviewed .The possible values for X are 0, 1,
2,and 3.The eight possible outcomes of this experiment are
 MMM , MMF , MFM , FMM , MFF , FMF , FFM , FFF  Where M denotes a male
candidate and F denotes a female candidate. If a candidate is equally likely to be a male
or a female, then each of the eight possible outcomes listed above is equally likely to
occur, and the probability distribution of X is:

x 0 1 2 3
P(X=x .125 .375 .375 .125
)

Expectation, E (X)

Experimental Approach

1
Suppose we throw an unbiased die 120 times and record the results in a frequency
distribution:

Score, x 1 2 3 4 5 6

Frequency, f 15 22 23 19 23 18

We can calculate the mean score obtained as follows

x
 fx  15  44  69  76  115  108  3.558
f 120

Theoretical Approach

The probability distribution for the random variable X, where X is a ‘the number on the
die’, is shown:

Score, x 1 2 3 4 5 6

P (X=x) 1 1 1 1 1 1
6 6 6 6 6 6

We can obtain a value for the ‘expected’ mean by multiplying each score by its
corresponding probability and summing.

1 1 1 1 1 1
Expected mean = 1   2   3   4   5   6 
6 6 6 6 6 6

21
= 3.5
6
If we have a statistical experiment
A practical approach results in a frequency distribution and a mean value,

A theoretical approach results in a probability distribution and an expected value, known

as the expectation.

The expectation of X (or expected value) written E (X), is given by

E (X) =  xP( X
allx
 x)

This can also be written

E (X) =  xi pi i= 1, 2, 3, . . . . .,n

Variance, Var(X)

Experimental Approach

For a frequency distribution with mean x , the variance, s2 is given by

2
s =
 f ( x  x) 2

. This can also be written s = 2  fx 2

 (x) 2
f f
Theoretical approach

For a discrete random variable X, with E (X) =  , the variance is defined as follows

The variance of X, written Var (X), is given by

Var (X) = E(X -  ) 2

2
Alternatively, Var(X) = E(X -  ) 2

= E (X2 - 2 X   2 )

= E(X2) – 2  E(X) + E (  2 )

= E (X2) - 2  2   2
= E (X2) -  2

So we have Var (X) = E (X2) -  2

Var(X) = E(X2) -  E ( X ) 2

Example # 1

The following table contains the probability distribution of the number of traffic
accidents in a small city.

Number of Accidents Daily (x) 0 1 2 3 4 5

P(X=x) 0.10 0.20 0.45 0.15 0.05 0.05

(a) Compute the mean or expected number of accidents per day.

(b) Compute the standard deviation.

Solution

( a) E ( X )   x P( X  x)
all x

 0  0.10  1 0.20  2  0.45  3  0.15  4  0.05  5  0.05

2
Var ( X )  E ( X 2 )   E ( X ) 
2
(b )

E ( X 2 )   x 2 P( X  x)
all x

 02  0.10  12  0.20  22  0.45  32  0.15  42  0.05  52  0.05

 0  0.20  1.8  1.35  0.8  1.25
 5.4
Var ( X )  E ( X 2 )   E ( X )
2

 5.4  (2) 2
 1.4
s.d ( X )  Var ( X )
 1.4
 1.18
BINOMIAL DISTRBUTION

James Bernoulli discovered binomial distribution in the year 1700. Let a random
experiment be performed repeatedly and let the occurrence of an event in a trial be called
a success and its non- occurrence is a failure. Consider a set of n independent Bernoulli
trials (n being finite), in which the probability ‘p’ of success in any trial is constant for
each trial. Then q = 1 – p, is the probability of failure in any trial.
The probability of x successes and consequently (n - x) failures in n independent trials, in
a specified order (say) SSFSFFFS. . . . . . FSF (Where S represents success and F failure)
is given by the compound probability theorem by the expression.

3
P(SSFSFFFS. . . . . .FSF) = P(S) P(S) P(F) P(S) P(F) P(F) P(F) P(S). . . . . . P(F) P(S) P(F)

= p. p .q . p .q .q .q .p. . . . . .q . p . q

= p. p. . . . .p q . q . q. q. . . . . q
(x factors) ((n-x) factors)
= p x q n-x
But x successes in n trials can occur in nCx ways and the probability for each of these
ways is p x q n-x. Hence the probability of x successes in n trials in any order whatsoever is
given by the addition theorem of probability by the expression:
n
Cx px qn-x
The probability distribution of the number of successes, so obtained is called the
Binomial probability distribution for the obvious reason that the probabilities of 0, 1,
2. . .n successes, viz.,
qn, nC1q n-1p , nC2 q n –2 p2 , . . . . . . ,p n, are the successive terms of the binomial expansion
(q + p)n

In general:

If the probability that an experiment results in a successful outcomes is p and the

probability that the outcome is a failure is q, where q = 1 – p, and if X is the r. v. ‘the
number of successful outcomes in n independent trails’, then the p d. f. of X is given by

P (X=x) = n C x p x q n-x x = 1, 2, 3, . . . . . . , n

The two independent constants n and p in the distribution are known as the parameters of
the distribution

Binomial distribution is discrete distribution as X can take only the integral values, viz.,
0, 1, 2, . . . . . , n. Any variable which follows binomial distribution is known as binomial
variate.

We shall use the notation X  B (n, p) to denote that the random variable X follows a
binomial distribution with parameters n and p.

Physical conditions for Binomial Distribution

We get the binomial distribution under the following experimental conditions.

(i) Each trial results in two mutually disjoint outcomes, termed as success and
failure
(ii) The number of trials ‘n’ is finite.
(iii) The trials are independent of each other
(iv) The probability of success ‘ p’ is constant for each trial.

The problems relating to tossing of a coin or throwing a dice or drawing cards from a
pack of cards with replacement lead to binomial probability distribution.
Expectation and Variance

If the random variable X is such that X  B (n, p)

Then E (X) = np
And Var (X) = npq, where q = 1- p

Example # 2

Suppose that the warranty records show the probability that a new computer needs a
warranty repair in the first 90 days is 0.05. If a sample of 10 new computers is selected,

4
(a) What is the probability that:

(i) none needs a warranty repair?

(ii) at least two needs a warranty repair?
(iii) more than two needs a warranty repair?

(b) What are the mean and standard deviation of the probability distribution?

Solution

(a) n = 10, p = 0.05, q = 0.95

(i) P(X = 0) = 0.599

(ii) P( X  2)  1   P( X  0)  P( X  1) 

= 1 – (0.599 +0.315)

= 1 - 0.914

= 0.086

(iii ) P ( X  2)  1   P( X  0)  P( X  1)  P( X  2) 
 1   0.599  0.315  0.075
 1  0.989
 0.011

(b) E ( X )  np

 10  0.05

 0.5
Var ( X )  npq
 10  0.05  0.95
 0.475
s.d ( X )  Var ( X )
 0.475
 0.69

POISSON DISTRIBUTION

Poisson Process

A Poisson process is said to exist if we can observe discrete events in an area of

opportunity - a continuous interval (of time, length, surface area etc.) – in a space manner
that if we shorten the area of opportunity or interval sufficiently,

1. the probability of observing exactly one success in the interval is stable

2. the probability of observing more than one success in the interval is 0.
3. the occurrence of a success in any one interval is statistically independent of
that in any other interval.

5
To better to understand Poisson process, suppose we examine the number of arriving
during 12 noon to 1 p.m. lunch hour at a bank located in the central business district in a
large city. Any arrival of a customer is a discrete event at a particular pointing time over
the continuous 1- hour interval. Over such an interval of time, there might be an average
of 180 arrivals. Now if we were to break up the 1- hour interval into 3600 consecutive 1-
second intervals,

The expected number (or average) number of customers arriving in any 1- second
interval would be 0.05

The probability of having more than one customer arriving in any one-second interval
approaches 0.

The arrival of one customer in any 1- second interval has no effect on the arrival of any
other customer in any other1- second interval.

The Poisson distribution has one parameter, called  , which is the average or expected
number of successes per unit. Interestingly, the variance of a Poisson distribution is also
equal to  and the standard deviation is equal to  . Moreover, the number successes x
of the Poisson random variable ranges from 0 to 
The mathematical expression for the Poisson distribution for obtaining x successes, given
that  successes are expected, is

e   x
P (X = x) = X = 0, 1, 2, 3. . . . . 
x!

Where,
 = mean number of occurrences (successes) in particular interval of time
e = mathematical constant approximated by 2.718
x = number of occurrences (successes)

If X is distributed in this way, then X ~ Po(  )

Examples of events, which might follow a Poisson distribution:

(i) the number of flaws in a given length of material

(ii) the number of accidents in a factory in one week.
(iii) The number of telephone calls made to a switch board in a given minute
(iv) The number of insurance claims made to a company in a given time
(v) The number of faulty blades in a pack of 100.
(vi) The number of defective material in packing manufactured by a good concern.
(vii) The number of printing mistakes at each page of the book
(viii) The number of cars passing a crossing per minute during the busy hours of a
day.

Example # 3

If, on average, three customers arrive per minute at the bank during the lunch hour, what
is the probability that in a given minute exactly two customers will arrive? Also, what is
the chance that more than two customers will arrive in a given minute?

Solution

6
Mean ( )  3

P ( X  2)  0.2240

P ( X  2)  1   P( X  0)  P( X  1)  P( X  2) 
 1   0.0498  0.1494  0.2240
 1  0.4232
 0.5768