Statistics For Management Note

CHAPTER THREE
PROBABLITY DISTRIBUTION

A probability distribution is the lists of all the possible values that a random variable can
take along with their probabilities.

Random variables
A variable that contains the outcome of a chance experiment
Example;
Suppose you want to find out the probability distribution for the number of heads on two
toss of a coin.

First toss Second toss

T T
T H
H H
H T

The probability distribution of number of heads

Number of Tosses Probability P(X)

Heads(x)

0 (T, T) 0.25
1 (T, H) + (H, T) 0.50
2 (H, H) 0.25
 P( X ) =1.00

Type of random variables

A. Discrete random variables;

▪A random variable is a discrete random variable if the set of all possible values is at
most a finite.
▪In most statistical situations, discrete random variables produce values that are non
negative whole numbers.

Example;
If six people are randomly selected from a population and how money of the six are
left-handed is to be determined, the random variable produced is discrete. The only
possible numbers of left-handed people in the sample of six are 0, 1, 2, 3, 4, 5,
and 6.There can not be 2.75 left handed people.
Other example; Determine the number of defects in a batch of 50 items.

Solomon M. 1
 Statistics For Management Note

B. Continuous random variables

▪ Variables that take on values at every point over a given interval.

Example;
▪If a person is assembling a product component; the time it takes to accomplish that
feat could be any value with in a reasonable range such as 3 minutes 36.4218 seconds
or 5 minutes 17.5169 seconds.
The outcome for random variables and their associated probabilities can be organized
in to distribution. The two type of distribution are;

1. Discrete distribution;
constructed from discrete random variables
2. Continuous distribution;
Based on continuous random variables.

Discrete Probability Distribution

Include the binomal distribution, Poisson distribution, and hyper geometric
distribution.

Mean or Expected Value of a Discrete Distribution

The mean or expected value of a discrete distribution is the long –run average of
occurrence.

 =  ( x) =
  x. p ( x)
Where,   x  =long run average
X=an out come
P(X) =the probability of that outcome

▪ Computing the mean for the tossing of the coin data.

X P(X) X.P(X)
0 0 .25 0.00
1 0.5 0.50
2 0.25 0.5
  X .P( X )
=1
 1 Heads

Solomon M. 2
 Statistics For Management Note

Standard deviations of a discrete distribution

=  ( x   ) 2 . p( x)

X P(X) X  2 X    2 .P  X 
0 0.25 0  0.25  2 =0.0625 0.0156
1 0.50 1  0.50  2 = 0.25 0.125
2 0.25 1  0.25  2 = 0.5625 0.141
 X    .P  X  =0.282
2

 2 = X   2 .P X  =0.282

 0.282 =0.531

1.2 Type of probability distribution

Binomial Distribution
Assumption;
The experiment involves n identical trials.
Each trials has only two possible outcomes denoted a success or as failure.
Each trial is independent of the previous trials
The term p and q remains constant through out the experiment. Where the term p is the
probability of getting a success on any one trials and the term q= (1-p) is the probability
of getting a failure on any one trail.

As the word binomial indicates, any single trial of a binomial experiment contains only
two possible outcomes. These two outcomes are labeled as success or failure. Usually the
outcome of interest to the researcher is labeled a success.

A binomial distribution is a discrete distribution. In n trials, only x success are possible,

where x is a whole number between 0 and n .For example, if five parts are randomly
selected from a batch parts, only 0, 1, 2, 3 , 4, or 5 defective parts are possible in
that sample.

In a binomial experiment, the trial must be independent. This constraint means that the
experiment is conducted with replacement. The effect of the independent trail

Solomon M. 3
 Statistics For Management Note

requirement is that p, the probability of getting a success on one trail, remains constant
from trail to trail.

If the sample size, n, is less than 5% of the population, the independence assumption is
not of great concern. Therefore the acceptable sample size for using the binomial
distribution with samples taken with out replacement is;

n < 5%N

Where n= Sample size

N= Population size

BINOMIAL FORMULA

n!
P(X) = C x p X .q n x = .p x .q n x
n
x!( n  x)!

Where;
n= the number of trials (number of being sampled)
x= the number of success desired
p= the probability of getting a success in one trail
q= the probability of getting a failure in one trail
Example;
A study conducted by the Gallup organization found that 65% of all financial consumers
are very satisfied with their primary financial institution. Suppose 40 financial consumers
are sampled randomly. What is the probability that exactly 23 of the 40 are very satisfied
with their primary financial institution?

Solution;
P=0.65 n=40
Q=1-p=0.35 x=23

P(X) = n C X . p X .q n  x = ( 40 C 23 ).(0.65) 25 .(0.35)17 =0.784

If 65% of the financial consumers are very satisfied, about 7.84% of the time the
researcher would get exactly 23 out of 40 financial consumers who are satisfied with their
financial institution.

Mean and Standard Deviation of a Binomial Distribution

Solomon M. 4
 Statistics For Management Note

  n. p
= n. p.q

A binomial distribution has an expected value or long run-average which is denoted by



Example;
According to one study 64% of all financial consumers believe banks are more
competitive today than they were five years ago. If 23 financial consumers are selected
randomly, what is the expected number who believe banks are more competitive today
than they were five years ago?

Solution;
n=23,
p=0.64
  n.p= (23).(0.64)
 = n. p.q = (23).(0.64).(0.36) =2.30
Poisson distribution

The Poisson distribution and the binomial distribution have some similarities, but also
have several differences.

The Poisson distribution focuses only on the number of discrete occurrences over some
interval or continuum. A poison experiment does not have a given number of trails (n) as
a binomial experiment does.For example, where as a binomial experiment might be used
to determine how many Ethiopian made car there are in a random sample of 20 cars, a
poison experiment might focus on the number of cars randomly arriving at an automobile
repair facility during a 10 minute interval.

The Poisson distribution describes the occurrence of rare events. For example, serious
accident at a chemical plant is rare, and the number might be described by the Poisson
distribution.

The Poisson distribution is often is used to describe the number of random arrival per
some time interval. If the number of arrivals per interval is too frequent, the time interval
can be reduced enough so that a rare number of occurrences is expected. Another
example of a Poisson distribution is the number of random customer arrivals per 5
minutes interval at a small boutique on weekday mornings.

The Poisson distribution has the following characteristics;

▪It is discrete distribution.
▪It describes rare events.
▪Each occurrence is independent of the other occurrences.

Solomon M. 5
 Statistics For Management Note

▪It describes discrete occurrences over a continuum.

▪the occurrence in each interval can range from zero to infinity.
▪The expected number of occurrences must hold constant through out the experiment

Example of Poisson –type situation include the following;

1. Number of telephone calls per minute at a small business.
2. Number of cases of a rare blood disease per 100,000 people.
3. Number of time a 1 year old personal computer printer break down per quarter (3
months).

Each of these examples represents a rare occurrence of events for some interval.
▪If a Poisson –distributed phenomenon is studied over a long period of time, a long run
average can be determined.
Poisson Formula

Is used to compute the probability of occurrences over an interval over a given lambda
value (  ).

 X e 
P(X) =
x!

Where;
x=0, 1, 2, 3…
 =long-run average
e= 2.718282
Here, x is the number of occurrences per interval for which the probability is being
computed,  is the long-run average, and e=20718282 is the base of natural logarithms.
The  value must hold constant through out a Poisson experiment.

Working Poisson Problems by Formula

Suppose bank customers arrive randomly on weekday afternoons at an average of 3.2

customers every 4 minutes. What is the probability of exactly five customers arriving in 4
minutes intervals on a weekday afternoon?

Solution;
 =3.2 customer/4 minutes
x=5 customers/4 minutes
P(x=5/  =3.2) =?

(3.2 5 )(2.718282) 3.2

P(x=5/  =3.2) =
5!

Solomon M. 6
 Statistics For Management Note

(335.54)(0.0408)
= =0.1141
120
What is the probability of having more than seven customers in 4 minute interval on a
weekday afternoon for the above example?
Solution;

 =3.2 customer/4 minutes

x>7 customers/4 minutes

The exact probabilities are summarized to find x>7

(3.2 8) (e) 3.2
P(x =8/  =3.2) = =0.111
8!

(3.2 9 ) (e) 3.2

P(x =9/  =3.2) = =0.040
9!
(3.210 ) (e) 3.2
P(x =10/  =3.2) = = 0.0013
10!
(3.211 ) (e) 3.2
P(x =11/  =3.2) = = 0.0004
11!
(3.212 ) (e) 3.2
P(x =12/  =3.2) = =0.0001
12!
13 3.2
P(x =13/  =3.2) = (3.2 ) (e) =.000
13!

P(x>7)=P(x≥8)=0.0169
It is unlikely that more than seven people would randomly arrive in any one 4-minute
period. This answer indicates that more than 7 people would randomly arrive in a 4-
minute period only 1.69% of the time .Bank officers could use these results to help them
to make staffing decision.

What is the probability of getting exactly 10 customers during an 8-minute interval?

Solution;
 =3.2 customer/4 minutes
x=10 customers/8 minutes

Note
*Never adjust or change x
*Always adjust the lambda

Solomon M. 7
 Statistics For Management Note

Logically, if the bank averages 3.2 customers every 4 minutes, it should average twice
as many or 6.4 customers every 8 minutes. If x were for a 2 minutes intervals, the value
of  would be halved from 3.2 to 1.6 customer per two minutes interval.
  =6.4 customers/ 8 minutes
x= 10 customers/ 8 minutes

10 6.4
 P(x =10/  =6.4) = (6.4) (e) =0.0528
10!

Mean and Standard Deviation of a Poisson distribution

*The mean or expected value of a Poisson distribution is  .

*The variance of a Poisson distribution is also  .
*The standard deviation of a Poisson distribution is  .

Approximating Binomial Problems by the Poisson distribution.

Certain type of binomial distribution problem can be approximated by using the

Poisson distribution. Binomial problem with large sample size and small p value, which
then generate rare events are potential candidate for use of the Poisson distribution.

As a rule of thumb, if n>20 and n.p≤7, the approximation is close enough to use the
Poisson distribution for binomial problems.
Example;
The following binomial distribution problem can be worked by using the Poisson
distribution.
n=5 and p = 0.03
*What is the probability that x = 4? That is;
P(x = 4/n =50) and p = 0.03 =?

 =  = n.q = (50) (0.03) =1.5

*As n>20 and n.p ≤ 7, this problem is candidate for the Poisson approximation.

4 1.5
 P(x =4/  =1.5) = (1.5 )(e ) = 0.0471
4!
*Using binomial formula yields
 ( 50 C 4 )(0.03) 4 (0.97) 46 =0.0459

 (0.0459 — 0.0471) = 0.0012 difference

Hyper geometric distribution

▪ Applies only to experiment in which the trial are done with replacement.

Solomon M. 8
 Statistics For Management Note

▪ consists of two possible outcomes; success and failure. However, the user must know
the size of the population and the proportion of success and failure in the population to
apply the hypergeometric distribution.That is, because the hypergeometric distribution is
used when sampling is done without replacement, information about population make up
must be known inorder to redetermine the probablity of a success in each successive trails
as the probablity changes.

The hypergeometric distribution has the followingcharacterstics.

▪ It is descrite distribution
▪ Each outcome consist of either a success or a falure
▪ Sampling is done with out replacement
▪ The population, N, is finite and known
▪ The number of success in the population,A, is known

Hyper geometric formula

C X . N  AC n x
P  x =
A

N Cn

Where;
N= Size of the population
n= Sample size
A= Number of success in the population
X= Number of success in the sample, Sampling is done with out replacement.

Hyper geometric distribution

•A distribution of probablities of the occurance of x items in a sample of n when there are
A of that same item in a populationof N.

Note;
The hypergeometric distribution should be used instead of the the binomial
distribution when;
1. Sampling is being done without replacement and
2. n  5%N

▪Hypergeometric probablities are calculated under the assumption that there is equaly
likely sampling of the remaining elements of the sample space.

Example;
▪ 24, of whom eight are woman, have applied for a job. If five of the applicants are
sampled randomly, what is the probablity that exactly three of those sampled are woman?

Solomon M. 9
 Statistics For Management Note

Solution;
N = 24
n=5
• The sample size is 21% of the population, which is  5% of the population (
n 5
  21% ).
N 24
• The populatioon brreakdown is;
 A= 8 women (success) and N–A = 24= 16 men
•The probablity of getting x = 3 women in the sample of n = 5 is

C 3 .16 C 2 (56)(120)
8
  0.18511
24 C 5 42,504

Solomon M. 10