ISL - 213E

STATISTICS - I
Istanbul Technical University
Faculty of Management
Department of Management Engineering
 Lecturer: Prof. Dr. Raziye Selim
Contact: [email protected]

Class Assistants:
Res. Assis. İrem Betül Koçak [email protected]

2
WEEK 11

Binomial, Negative Binomial,

Hypergeometric and Poisson
Distributions
3
 The Binomial Probability Distribution

 A probability distribution is called theoretical when the

mathematical properties of its random variable are used to
produce its probabilities.

 The simplest theoretical probability distribution we will

consider is one that describes processes whose trials have only
two possible outcomes.

4
 The Binomial Probability Distribution

Examples:

• A quality contol system in a manufacturing plant labels each tested

item as either defective or acceptable

• A firm that sales car, when show a car to customers;

They can buy or not buy

• The personal manager in an organization is faced with a two-stage

process each time he offers a job either the applicant accepts the offer
or rejects it

5
 Characteristics of a Discrete Probability Distribution

 The probability of x successes in n trials of a process which meets the

following conditions.

1) A trial has only two possible outcomes; a success or a failure

2) There is a fixed number, n of identical trials

3) The trials of experiment are independent of each other

6
 Characteristics of a Discrete Probability Distribution

4) The probability, p , associated with a success remains constant

from trial to trial

5) q: the probability of a failure

p+q=1

7
 Example

 Household secuity produces home security units for houses. If the company is
operating at standard quality, 10% of the security systems will have problems.
• There are only two possible outcomes when a unit is sold.
It is good or it is defective.
Finding a defective system in this application will be considered a success.
• The probability of a defective system, p=0,10
• The probablity of a good system q=1-p=0,90

Household Security analysed four security systems each week.

1) What’s the probability 0 defective?
2) What’s the probability exactly one defective?
3) What’s the probability two defective?
4) What is expected value and variance?

8
 Example

1) P(X=0 defectives) =
2) The probability of exactly one defective in a sample of four;
P(X=1) = = 0,2916
GGGD
GGDG
4 df situation
GDGG
3) P(X=2) = DGGG

P(X=3) = 0,0036
P(X=4) = 0,0001

4) E(X) = np = 4(0,10) = 0,4

V(X) = npq

9
 Example

 The catalog company sales manager had an 11% return rate goal.
She tracked a sample of 300 items and observed that 45 were returned.
The expected number of returned was;
E(X) = 300 * (0,11) = 33
(X) =

The binomial distribution is bell shaped

µ±1 68% of the outcomes
µ±2 95%
µ±3 99,7%
33 3(5,419)
16,74......................................49,26

10
 Example

 The sample of 300 items the manager tracked showed 45 returns. This is
within the range of possible outcomes. But it is located in the upper tail of
the distribution. The manager would have cause for concern that the 11%
return rate has increased.

11
 Example

 The 1997 Tenth Planet Teachers Survey reported that, 21% of

elementary teachers use the web.

If 5 teachers are selected at random, what’s the probability that;

a) Exactly 3 of the teachers use the web

b) Less than 4 teachers use the web.

c) More than 1 teacher uses the web.

 p = 0,21 n = 5

12
 Example
 p = 0,21 n = 5
a) Exactly 3 of the teachers use the web
P(X=3) =

b) Less than 4 teachers use the web.

P(X<4) = 1 – P(X4)
= 1 – [P(4) + P(5)]
= 1 – [0,0077 + 0,0004]
= 0,992
c) More than 1 teacher uses the web.
P(X>1) = 1 – [P(X=0) + P(X=1)]
= 1 – [0,7166]
= 0,2834

13
 Example 4.50

 p = 0,37 n=8

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

= 0,0248 + 0,1166 + 0,2397 + 0,2815

= 0,6626

c) P(X>2) = 1 – [P(0) + P(1)] = 1 - [0,0248 + 0,1166] = 0,8586

14
 Example

 Suppose that the insurance broker, Shirley Ferguson, has five contracts and she
believes that for each contract, the probability of making a sale is 0,40. Find;

a) The probability that she makes at most one sale.

b) The probability that she makes between two or four sales.

c) Graph the probability distribution function.

15
 Example

a) The probability that she makes at most one sale.

P(X1) = P(0) + P(1) = 0,078 + 0,259 = 0,337

b) The probability that she makes between two or four sales.

P(2X4) = P(2) + P(3) + P(4) = 0,346 + 0,230 + 0,077 = 0,653

c) Graph the probability distribution function.

Making a
P(X)
Sale
0 0,0778
1 0,2592
2 0,3456
3 0,2304
4 0,0768
5 0,0102
1,0000

16
 Negative Binomial (Geometric Distributions)
(Pascal Distribution)

 We are sometimes interested in the number of the trial on which the kth
success occurs.

 For instance, we maybe interested in;

• the probability that the tenth child exposed to a contagious disease will
be third to catch it;
• the probability that the fifth person to hear a rumor will be first one to
believe it.
• the probability that a burglar will be caught for the second time on his
or her eighth job.

17
 Negative Binomial (Geometric Distributions)
(Pascal Distribution)

 If the kth success is to occur on the xth trial, there must be k-1 successes
on the first x-1 trials, the probability for this is;

 The probability of a success on the xth trial is p, and the probability that
the kth success occurs on the xth trial is, therefore,

x has a negative binomial distribution (waiting time dist.)

18
 Example

Example: If the probability is 0,40 that a child exposed to a certain contagious

disease will catch it, what’s the probability the tenth child exposed to the disease
will be the third to catch it?

x = 10
k=3 ()
9 ∗ 0 , 40 3 ∗ 0 ,60 7=0,0645
2
p = 0,40

Mean: =

k=1 the name is geometric distribution

g(x,p) = p for x=1,2,3

19
 Example

Example: If the probability is 0,75 that an applicant for a driver’s license will pass the
road test on any given try, what’s the prob. that an applicant will finally pass the test
on the fourth try?
x=4
p = 0,75

g(4 ; 0,75) = 0,75 = 0,75 = 0,0117

20
 Hypergeometric Distribution

 Sample data are selected without replacement, from a finite population. Thus,
the outcome of one observation is dependent on the outcomes of the previous
observations.

 N elements of which k are looked upon as success and the other N-k as
failures.

 As in connection with the binomial dist, we are interested in the probability of

getting x success in n trials, but we are choosing, without replacement, n of the
N elements contained in the set.

21
 Hypergeometric Distribution

 There are ways of choosing x of the k successes and

ways of choosing n-x of the N-k failures,

and hence ways of choosing x successes and n-x failures.

«x successes in n trials»

Hyp(N, k, n)

22
 Example

 As there are 20 balls in a bag, 8 of them are white.

If we draw 5 without replacement;
E(X) = = 2
X P(x)
0 0,05108
1 0,25542
2 0,39732 V(X) = npq
3 0,23839
4 0,05408 =
5 0,00361
= * 5 * 0,40 * 0,60

= 0,94737

23
 Example

 As part of an air-pollution survey, an inspector decides to examine the

exhaust of six of a company’s 24 trucks. If 4 of the company’s trucks emit
excessive amounts of pollutants, what is the probability that none of them
will be included in the inspector’s sample?

x=0
n=6
0,2880
N = 24
k=4

24
 Example

 Among the 120 applicants for a job, only 80 are actually qualified. If 5 of the
applicants are randomly selected for an in-depth interview, find the probability that
only 2 of the 5 will be qualified for the job.
Hypergeometric
a) x = 2, n = 5, N = 120, k = 80

h(2 ; 5 ; 120 ; 80) ¿ 0,164

¿( )( ) (
5 2 2 3

)
Binomial
b) b(2 ; 5 ; 2/3) 1 −2 0,165
2 3 3

25
 Poisson Distribution

 Many studies are based on counts of the times, a particular event

occurs in a given area of opportunity.

 An area of opportunity is a continuous unit or interval of time, volume

or any physical area in which there can be more than one occurence of
an event.

 Ex:
• Surface defects on a new refrigerator
• The number of network failures in a day
• The number of people arriving at a bank

26
 Poisson Distribution

 We are interested in counting the number of times a particular event occurs in a

given area of opportunity. The area of opportunity is defined by time, length,
surface area.

 The probability that an event occurs in a given area of opportunity is the same for
all the areas of opportunity.

 The number of events that occur in one area of opportunity is independent of the
number of events that occur in any other area of opportunity

 The prob. that two or more events will occur in an area of opportunity approaches
zero as the area of opportunity becomes smaller.
27
 Poisson Distribution

 Consider the number of customers arriving during the lunch hour

at a bank. You are interested in the # of customers that arrive each
minute.

: lambda = expected number of events per unit.

−λ x
ⅇ λ
p(x) =
x!

28
 Example

 Suppose ;
a) What’s the probability that in a given minute, exactly two customers will
arrive?
ⅇ− 3 32 9
p(x=2) = = 3
=0,2240
2! 2,71828 ∗2

b) What’s the probability that more than two customers will arrive in a given
minute?

p(x>2) = p(x=3) + p(x=4) + ...... + p(x=)

= 1 – p(x≤2)
= 1 – [p(x=0)+p(x=1)+p(x=2)]
= 1 – (0,0498 + 0,1494 + 0,2240) = 0,5678
57,68% chance that more than two customers will arrive in the same minute.

29
 Z= x’s independent random variables having the
same Bernoulli dist. success, failure
E(Z) = p Var(Z) = p(1-p)
𝐩 ( 𝐱 )= p x ( 1− p )1− x

 When n is large, the calculation of binomial probabilities with the formula as follow;

Consider a random variable Y=X/n , where X is a random variable having

a binomial dist. with the parameters n and p. This (Y) random variable is
the proportion of success in n trials.

E(Y) = E(X/n) = V(Y) = V(X/n) =

Will usually involve a prohibitive amount of work.

n∞, p0 np remains constant np = p=

30
 Write in binomial formulea as follow;

( )( ) ( )
x n− x
n λ λ
1−
x n n

ⅇ− λ λ x Poisson
We will find p(x ; ) = (French mathematician
x! Simeon Poisson 1781-1840)

• If n20 and p≤0,05 a good approximation

• When n100 and np<10 excellent approximation

31
 Example

 If 2% of the books bound at a certain bindery(ciltevi) have defective bindings,

what’s the prob. that 5 of 400 books will have defective bindings?

n = 400 x=5 = np = 400 (0,02) = 8

ⅇ− 8 85
( )
400 ( 0 , 02)5 0 , 98 395
5
p (5 ; 8)=
5!
=0,093

32
 Example

 The prob. is 0,00005 that a car will have a flat tire while crossing a certain bridge.
Among 10.000 cars crossing this bridge;
a. Exactly two will have a flat tire
b. At most two will have a flat tire

p = 0,00005 n = 10.000 = np = 0,5

a. x=2
ⅇ −0 , 5 0 , 52
p ( 2 ; 0 , 5 )= =0,0758
2!
b. p(x=0) + p(x=1) + p(x=2)

ⅇ− 0 ,5 0 , 50 ⅇ− 0 ,5 0 ,51 ⅇ− 0 ,5 0 , 52
+ + =0,6065+0,3033 +0,0758=0,9856
0! 1! 2!
E(X) = Var(X) =
33
 THANKS!

Any questions?


34
 Presentation Design Credits

Presentation Template
 Catalina, Jimena. (2021). Salerio Powerpoint template.
Retrieved from https://www.slidescarnival.com/salerio-free-presentation-template/1875