Applied Business Statistics, 7th ed.

by Ken Black

Chapter 5
Discrete
Distributions

Copyright2011
Copyright 2011John
JohnWiley
Wiley&&Sons,
Sons,Inc.
Inc. 1
 Learning Objectives

Distinguish between discrete random variables and

continuous random variables.
Know how to determine the mean and variance of a
discrete distribution.
Identify the type of statistical experiments that can be
described by the binomial distribution, and know how to
calculate probabilities based on the binomial distribution.

Copyright 2011 John Wiley & Sons, Inc. 2

 Discrete vs. Continuous Distributions

Random Variable - a variable which contains the outcomes of a chance

experiment
Discrete Random Variable – A random variable that only takes on distinct values
ex: Number of heads on 10 flips, Number of defective items in a random sample of 100,
Number of times you check your watch during class, etc.
Continuous Random Variable – A random variable that takes on infinite values
by increasing precision. For each two values, there always exists a valid value in
between them.
ex: Time until a bulb goes out, height, etc.

Copyright 2011 John Wiley & Sons, Inc. 3

 Describing a Distribution

A distribution can be described by constructing a

graph of the distribution
Measures of central tendency and variability can be
applied to distributions

Copyright 2011 John Wiley & Sons, Inc. 4

 Describing a Discrete Distribution

Mean of discrete distribution – is the long run average

If the process is repeated long enough, the average of the outcomes will approach
the long run average (mean)
Mean of a discrete distribution
µ = ∑ (Xi * P(Xi))
where µ is the long run average,
Xi = the ith outcome

Copyright 2011 John Wiley & Sons, Inc. 5

 Describing a Discrete Distribution

Variance of a discrete distribution is obtained in a manner

similar to raw data, summing the squared deviations from
the mean and weighting them by P(Xi) (rather than dividing
by n):
Var(Xi) = ∑ ((Xi – )2* P(Xi))
Standard Deviation is computed by taking the square root of
the variance

Copyright 2011 John Wiley & Sons, Inc. 6

 Discrete Distribution -- Example

An executive is considering out-of-town business

travel for a given Friday. At least one crisis could occur
on the day that the executive is gone. The distribution
on the following slide contains the number of crises
that could occur during the day the executive is gone
and the probability that each number will occur. For
example, there is a .37 probability that no crisis will
occur, a .31 probability of one crisis, and so on.

Copyright 2011 John Wiley & Sons, Inc. 7

 Discrete Distribution -- Example

Distribution of Daily Crises

P 0.5
Number of r
Probability o 0.4
Crises
b
a 0.3
0 0.37 b
1 0.31 i 0.2
2 0.18 l
i 0.1
3 0.09 t
4 0.04 y 0
5 0.01 0 1 2 3 4 5
Number of Crises

Copyright 2011 John Wiley & Sons, Inc. 8

 Mean and Standard Deviation
of a Discrete Distribution
# Crises (X) Prob (P) XP X- (X- (X-P
0 0.37 0 -1.15 1.32 0.49
1 0.31 0.31 -0.15 0.02 0.01
2 0.18 0.36 0.85 0.72 0.13
3 0.09 0.27 1.85 3.42 0.31
4 0.04 0.16 2.85 8.12 0.32
5 0.01 0.05 3.85 14.82 0.15
SUM: 1 1.15 8.1 28.44 1.41
   XP  1.15
  2   X    P  X   1.41  1.19
2

Copyright 2011 John Wiley & Sons, Inc. 9

 Requirements for a Discrete
Probability Function -- Examples
- Each probability must be between 0 and 1
- The sum of all probabilities must be equal to 1.
X P(X) X P(X) X P(X)

-1 .1 -1 -.1 -1 .1
0 .2 0 .3 0 .3
1 .4 1 .4 1 .4
2 .2 2 .3 2 .3
3 .1 3 .1 3 .1
1.0 1.0 1.2
VALID NOT NOT
VALID VALID
Copyright 2011 John Wiley & Sons, Inc. 10
 Binomial Distribution
The binomial distribution is a discrete distribution where X is the number of “successes” and the following
four conditions are met:
There are n trials
The n trials are independent of each other
The outcome is dichotomous – only two outcomes possible
The probability of “success” is constant
Example, 10 coin flips, X = # of heads
X = the number of “successes” and we say X follows a Binomial distribution with n trial and P(success) = p
If the data follow a binomial distribution, then we can summarize P(X i) for all values of Xi = 1, …, n through
the binomial probability distribution formula

Copyright 2011 John Wiley & Sons, Inc. 11

 Binomial Distribution
n! X n  X
Probability P( X )  p q
function X ! n  X !
for 0  X  n, q  1  p
Mean value
  n p
Variance and
Standard
Deviation  2  n pq
   2  n pq

Copyright 2011 John Wiley & Sons, Inc. 12

 Binomial Distribution:
Demonstration Problem 5.3
According to the U.S. Census Bureau,
approximately 6% of all workers in Jackson,
Mississippi, are unemployed. In conducting a
random telephone survey in Jackson, what is the
probability of getting two or fewer unemployed
workers in a sample of 20?

Copyright 2011 John Wiley & Sons, Inc. 13

 Binomial Distribution:
Demonstration Problem 5.3
In this example,
6% are unemployed => p
The sample size is 20 => n
94% are employed => q
X is the number of successes desired
What is the probability of getting 2 or fewer unemployed
workers in the sample of 20? => P(X≤2)
The hard part of this problem is identifying p, n, and x –
emphasize this when studying the problems.

Copyright 2011 John Wiley & Sons, Inc. 14

 Binomial Distribution:
Demonstration Problem 5.3
n  20
p  .06
q  .94
P ( X  2)  P ( X  0)  P ( X  1)  P ( X  2)
 .2901  .3703  .2246  .8850
20! 0 20  0
P ( X  0)   .06  .94  (1)(1)(.2901)  .2901
0!(20  0)!

20! 1 20  1
P( X  1)     
.06 .94  (20)(.06)(.3086)  .3703
1!(20  1)!

20! 2 20  2
P( X  2)   .06  .94  (190)(.0036)(.3283)  .2246
2!(20  2)!

Copyright 2011 John Wiley & Sons, Inc. 15

 Binomial Distribution:
Demonstration Problem 5.3

What are the mean and standard deviation of this

distribution?
  n  p  (20)(.06)  1.20
 2  n  p  q  ( 20)(.06)(.94)  1.128
   2  1.128  1.062

Copyright 2011 John Wiley & Sons, Inc. 16

 Binomial Distribution Table:
Demonstration Problem 5.3
n  20
n = 20 PROBABILITY
X 0.05 0.06 0.07 p . 06
0 0.3585 0.2901 0.2342 q . 94
1 0.3774 0.3703 0.3526 P( X  2 )  P( X  0 )  P( X  1)  P( X  2 )
2 0.1887 0.2246 0.2521 . 2901. 3703. 2246 . 8850
3 0.0596 0.0860 0.1139
4 0.0133 0.0233 0.0364
5 0.0022 0.0048 0.0088
6 0.0003 0.0008 0.0017
7 0.0000 0.0001 0.0002
8 0.0000 0.0000 0.0000
… … … …
20 0.0000 0.0000 0.0000

Copyright 2011 John Wiley & Sons, Inc. 17

 Excel’s Binomial Function

n = 20
p = 0.06

X P(X)
0 =BINOMDIST(A5,B$1,B$2,FALSE)
1 =BINOMDIST(A6,B$1,B$2,FALSE)
2 =BINOMDIST(A7,B$1,B$2,FALSE)
3 =BINOMDIST(A8,B$1,B$2,FALSE)
4 =BINOMDIST(A9,B$1,B$2,FALSE)
5 =BINOMDIST(A10,B$1,B$2,FALSE)
6 =BINOMDIST(A11,B$1,B$2,FALSE)
7 =BINOMDIST(A12,B$1,B$2,FALSE)
8 =BINOMDIST(A13,B$1,B$2,FALSE)
9 =BINOMDIST(A14,B$1,B$2,FALSE)

Copyright 2011 John Wiley & Sons, Inc. 18

 Minitab’s Binomial Function

X P(X =x)

0 0.000000
1 0.000000
2 0.000000
3 0.000001
4 0.000006
5 0.000037
6 0.000199
7 0.000858 Binomial with n = 23 and p = 0.64
8 0.003051
9 0.009040
10 0.022500
11 0.047273
12 0.084041
13 0.126420
14 0.160533
15 0.171236
16 0.152209
17 0.111421
18 0.066027
19 0.030890
20 0.010983
21 0.002789
22 0.000451
23 0.000035

Copyright 2011 John Wiley & Sons, Inc. 19

 Poisson Distribution

The Poisson distribution focuses only on the number

of discrete occurrences over some interval or
continuum
Poisson does not have a given number of trials (n)
as a binomial experiment does
Occurrences are independent of other occurrences
Occurrences occur over an interval

Copyright 2011 John Wiley & Sons, Inc. 20

 Poisson Distribution

If Poisson distribution is studied over a long period

of time, a long run average can be determined
The average is denoted by lambda (λ)
Each Poisson distribution contains a lambda value from
which the probabilities are determined
A Poisson distribution can be described by λ alone

Copyright 2011 John Wiley & Sons, Inc. 21

 Poisson Distribution

Probability function

 X e 
P( X )  for X  0,1,2,3,...
X!
where :
  longrun average
e  2.718282... (the base of natural logarithms)
 Mean value  Variance  Standard deviation

  
Copyright 2011 John Wiley & Sons, Inc. 22
 Poisson Distribution:
Demonstration Problem 5.7
Bank customers arrive randomly on weekday
afternoons at an average of 3.2 customers
every 4 minutes. What is the probability of
having more than 7 customers in a 4-minute
interval on a weekday afternoon?

Copyright 2011 John Wiley & Sons, Inc. 23

 Excel’s Poisson Function

= 1.6

X P(X)
0 =POISSON(D5,E$1,FALSE)
1 =POISSON(D6,E$1,FALSE)
2 =POISSON(D7,E$1,FALSE)
3 =POISSON(D8,E$1,FALSE)
4 =POISSON(D9,E$1,FALSE)
5 =POISSON(D10,E$1,FALSE)
6 =POISSON(D11,E$1,FALSE)
7 =POISSON(D12,E$1,FALSE)
8 =POISSON(D13,E$1,FALSE)
9 =POISSON(D14,E$1,FALSE)

Copyright 2011 John Wiley & Sons, Inc. 24

 Minitab’s Poisson Function

X P(X =x)

0 0.149569
1 0.284180
2 0.269971 Poisson with mean = 1.9
3 0.170982
4 0.081216
5 0.030862
6 0.009773
7 0.002653
8 0.000630
9 0.000133
10 0.000025

Copyright 2011 John Wiley & Sons, Inc. 25

 Poisson Distribution:
Demonstration Problem 5.7 - Solution
λ = 3.2 customers/4 minutes
We want to calculate P(X > 7 customers/4 minutes)
The problem can either be solved as:
P(X>7) = P(X=8) + P(X=9) + …, or
P(X>7) = 1 – P(X≤7) = 1 – [P(X=7) + P(X=6) + … + P(X=0)]
The answer can be obtained directly or through software
The answer you get is 1.7% of the time.
Bank officers could use these results to help them make
staffing decisions.

Copyright 2011 John Wiley & Sons, Inc. 26

 Hypergeometric Distribution

Sampling without replacement from a finite

population
The number of objects in the population is denoted N.
Each trial has exactly two possible outcomes, success
and failure.
Trials are not independent
X is the number of successes in the n trials
The binomial is an acceptable approximation,
if n < 5% N. Otherwise it is not.

Copyright 2011 John Wiley & Sons, Inc. 27

 Hypergeometric Distribution

Probability function  ACx   N  ACn  x 

N is population size P( x ) 
n is sample size
N Cn
A is number of successes in population
x is number of successes in sample
Mean An
Value  
N

Variance and standard A( N  A) n( N  n)


2
 2
deviation N ( N  1)

 
2


Copyright 2011 John Wiley & Sons, Inc. 28

 Hypergeometric Distribution:
Probability Computations
In a population of 8 males and 16 females, what is the chance of
choosing 3 males and 2 females?

Population Sample
Size 24 (N) 5 (n)
P ( x  3) 
 ACx  N  ACn  x 
Males 8 (A) 3 (x) N Cn
X P(X)

 8C 3  24  8C 5  3 
0 0.103
C5
24
1
2
0.343
0.369 
 56120 
3 0.158 42,504
4 0.026
 .158
5 0.001

Copyright 2011 John Wiley & Sons, Inc. 29

 Excel’s Hypergeometric Function

N = 24
A= 8
n= 5

X P(X)
0 =HYPGEOMDIST(A6,B$3,B$2,B$1)
1 =HYPGEOMDIST(A7,B$3,B$2,B$1)
2 =HYPGEOMDIST(A8,B$3,B$2,B$1)
3 =HYPGEOMDIST(A9,B$3,B$2,B$1)
4 =HYPGEOMDIST(A10,B$3,B$2,B$1)
5 =HYPGEOMDIST(A11,B$3,B$2,B$1)

=SUM(B6:B11)

Copyright 2011 John Wiley & Sons, Inc. 30

 Minitab’s Hypergeometric Function

X P(X =x)

0 0.102767
1 0.342556
Hypergeometric with N = 24, A = 8, n = 5
2 0.368906
3 0.158103
4 0.026350
5 0.001318

Copyright 2011 John Wiley & Sons, Inc. 31