Essentials of Statistics for

Business and Economics, 10e

Chapter 5 – Discrete Probability

Distributions

Camm, Cochran, Fry, Ohlmann, Anderson, Sweeney, Williams, Essentials of Statistics for Business &
Economics, 10th Edition. © 2024 Cengage Group. All Rights Reserved. May not be scanned, copied or
duplicated, or posted to a publicly accessible website, in whole or in part.
 Chapter Contents

5.1 Random Variables

5.2 Developing Discrete Probability Distributions
5.3 Expected Value and Variance
5.4 Bivariate Distributions, Covariance, and Financial Portfolios
5.5 Binomial Probability Distribution
5.6 Poisson Probability Distribution
5.7 Hypergeometric Probability Distribution
Summary

© 2024 Cengage Group. All Rights Reserved.

 Chapter Objectives (1 of 2)

After completing this chapter, you will be able to:

LO 5-1 Define and identify discrete and continuous random variables.
LO 5-2 Develop, graph, and interpret valid probability distributions for discrete
random variables.
LO 5-3 Calculate and interpret the expected value, variance, and standard deviation
for discrete random variables.
LO 5-4 Develop and interpret the bivariate probability distribution for two discrete
random variables.
LO 5-5 Calculate and interpret the covariance and correlation coefficient for two
discrete random variables.
LO 5-6 Calculate and interpret the expected value and variance of a linear
combination of two random variables.

© 2024 Cengage Group. All Rights Reserved.

 Chapter Objectives (2 of 2)

LO 5-7 Identify and use the binomial probability function to compute the probability of
a specified number of successes out of a given number of trials.
LO 5-8 Calculate and interpret the expected value, variance, and standard deviation
for binomial experiments.
LO 5-9 Identify and use the Poisson probability function to compute the probability of
a specified number of occurrences over a given interval.
LO 5-10 Calculate and interpret the expected value, variance, and standard deviation
for Poisson experiments.
LO 5-11 Identify and use the hypergeometric probability function to compute the
probability of a specified number of successes out of a given number of trials.
LO 5-12 Calculate and interpret the expected value, variance, and standard deviation
for hypergeometric experiments.

© 2024 Cengage Group. All Rights Reserved.

 Introduction
In this chapter, we extend the study of probability by introducing the concepts of random
variables and probability distributions.
Random variables and probability distributions are models for populations of data.
The values of random variables represent the values of the data, and the probability
distribution provides either the probability of each data value or a rule for computing the
probability of each data value or a set of data values
The focus of this chapter is on probability distributions for discrete data, that is, discrete
probabil­ity distributions.
We will introduce two types of discrete probability distributions.
• A table with one column for the values of the random variable and a second column
for the asso­ciated probabilities.
• Three types of mathematical functions that represent a probability distribution: the
binomial, Poisson, and hypergeometric distribution.

© 2024 Cengage Group. All Rights Reserved.

 5.1 Examples of Discrete Random Variables
A random variable is a numerical description of the outcome of an experiment.
A discrete random variable may assume either a finite number of values or an infinite
sequence of values.
Possible Values for the
Random Experiment Random Variable (x) Random Variable
Flip a coin Face of coin showing 1 if heads; 0 if tails

Roll a die Number of dots showing 1, 2, 3, 4, 5, 6

on top of die
Contact five customers Number of customers who 0, 1, 2, 3, 4, 5
place an order
Operate a health care Number of patients who 0, 1, 2, 3, …
clinic for one day arrive
Offer a customer the Product chosen by 0 if none; 1 if choose
choice of two products customer product A; 2 if choose
product B

© 2024 Cengage Group. All Rights Reserved.

 5.1 Examples of Continuous Random Variables

A continuous random variable may assume any numerical value in an interval or collection
of intervals (*see notes.)
Possible Values for the
Random Experiment Random Variable (x) Random Variable
Customer visits a web Time customer spends on x≥0
page web page in minutes
Fill a soft drink can (max Number of ounces 0 ≤ x ≤ 12.1
capacity = 12.1 ounces)
Test a new chemical Temperature when the 150 ≤ x ≤ 212
process desired reaction takes
place (min temperature
= 150°F; max
temperature = 212°F)
Invest $10,000 in the Value of investment after x≥0
stock market one year

© 2024 Cengage Group. All Rights Reserved.

 5.2 Discrete Probability Distributions

The probability distribution for a random variable describes how probabilities are distributed
over the values of the random variable.
We can describe a discrete probability distribution with a table, graph, or formula.

The Two Types of Discrete Probability Function

Probability Distributions The probability function, denoted by f(x),
1. Uses the rules of assigning probabilities defines the probability distribution by
to experimental outcomes to determine providing the probability for each value of the
probabilities for each value of the random random variable.
variable. The required conditions for a discrete
probability function are:
2. Uses a special mathematical formula to
compute the probabilities for each value
of the random variable.

© 2024 Cengage Group. All Rights Reserved.

 5.2 Empirical Discrete Distribution
There are three methods for assigning probabilities to x # of days f(x)
random variables: classical method, subjective method, 0 54 54 / 300 = 0.18
and relative frequency method. 1 117 117 / 300 = 0.39
2 72 72 / 300 = 0.24
The use of the relative frequency method to develop
3 42 42 / 300 = 0.14
discrete probability distributions leads to an empirical
4 12 12 / 300 = 0.04
discrete distribution. 5 3 3 / 300 = 0.01
Example: Number of automobiles sold during a day at 300 Total 1.00
DiCarlo Motors.
The table to the right shows the number of automobiles
sold at DiCarlo over the past 300 days.
We define the random variable of interest as:
= the number of automobiles sold during a day.
We can use the relative frequencies to represent
graphically the empirical discrete distribution for .

© 2024 Cengage Group. All Rights Reserved.

 5.2 Probability Distribution Given by a Formula
In addition to tables and graphs, a formula that gives the probability function, , for every value
of x is often used to describe the probability distributions.
Typical discrete probability distributions specified by formulas are the discrete-uniform,
binomial, Poisson, and hypergeometric distributions.
The discrete uniform probability distribution is the simplest example of a discrete
probability distribution given by a formula.
The discrete uniform probability function is

Where: n = the number of values the random variable may assume

An example of a discrete uniform probability distribution is provided by the experiment of rolling
a die, in which equally likely outcomes, the random variable is represented by the number of
dots on the upward face of the die, and .

© 2024 Cengage Group. All Rights Reserved.

 5.3 Expected Value

The expected value, or mean, of a random variable is a measure of its central location.
The expected value of a discrete random variable is calculated as a weighted average of the
values the random variable may assume. The weights are the probabilities.

x f(x) xf(x)
The expected value for the DiCarlo Motors example is 0 0.18 0(0.18) = 0.00
automobiles (*see notes) 1 0.39 1(0.39) = 0.39
2 0.24 2(0.24) = 0.48
Although sales from 0 to 5 automobiles are possible on 3 0.14 3(0.14) = 0.42
any one day, over time DiCarlo can anticipate selling 4 0.04 4(0.04) = 0.16
an average of 1.50 automobiles per day, or 30(1.50) = 5 0.01 5(0.01) = 0.05
45 automobiles per month.
1.50

© 2024 Cengage Group. All Rights Reserved.

 5.3 Variance and Standard Deviation

The variance summarizes the variability in the values of a random variable (*see notes.)
The variance of a discrete random variable is calculated as

For the DiCarlo Motors example, variance and standard deviation are calculated as

x x−μ (x − μ)2 f(x) (x − μ)2 f(x)

automobiles 0 0 − 1.5 = −1.50 2.25 0.18 2.25(0.18) = 0.4050
Because the standard deviation is measured in 1 1 − 1.5 = −0.50 0.25 0.39 0.25(0.39) = 0.0975
the same units as the random variable, it is 2 2 − 1.5 = 0.50 0.25 0.24 0.25(0.24) = 0.0600
often preferred in describing the variability of a 3 3 − 1.5 = 1.50 2.25 0.14 2.25(0.14) = 0.3150
random vari­able. 4 4 − 1.5 = 2.50 6.25 0.04 6.25(0.04) = 0.2500
5 5 − 1.5 = 3.50 12.25 0.01 12.25(0.01) = 0.1225
1.2500

© 2024 Cengage Group. All Rights Reserved.

 5.4 Bivariate Probability Distributions

A bivariate probability distribution is a probability distribution involving two random

variables.
The outcomes of a bivariate experiment consist of the values of two random variables.
Examples of bivariate experiments:
• Rolling a pair of dice and recording the number for each die.
• Recording the percentage gain for a stock fund and a bond fund over a period of time.
When dealing with bivariate probability distributions, we are often interested in the relationship
between the two random variables.
In this section, we introduce bivariate distributions and show how the covariance and
correlation coefficient can be used as a measure of linear association between the random
variables.
We shall also see how bivariate probability distributions can be used to construct and analyze
financial portfolios.

© 2024 Cengage Group. All Rights Reserved.

 5.4 A Bivariate Experiment for DiCarlo Motors
Suppose we consider the bivariate experiment of observing a day of operations at DiCarlo
Motors and recording the number of cars sold daily at the DiCarlo Motors automobile
dealerships in Saratoga and Geneva, New York.
Let us define, the number of cars sold at the Geneva dealership, and the number of cars sold
at the Saratoga dealership.
The table shows the number of cars sold (frequencies) at each of the dealerships over a 300-
day period.
Saratoga Dealership
Geneva Dealership 0 1 2 3 4 5 Total
0 21 30 24 9 2 0 86
1 21 36 33 18 2 1 111
2 9 42 9 12 3 2 77
3 3 9 6 3 5 0 26
Total 54 117 72 42 12 3 300

© 2024 Cengage Group. All Rights Reserved.

 5.4 A Bivariate Empirical Discrete Probability
Distribution for DiCarlo Motors
If we divide all the frequencies by the 300 observations, we develop a bivariate empirical
discrete probability distribution for automobile sales at the two DiCarlo dealerships.
The probabilities in the lower and right margins provide the marginal distributions for the
DiCarlo Motors Saratoga and Geneva dealerships, respectively.
The probabilities in the table provide the bivariate probability distribution for sales at both
dealerships.
Saratoga Dealership
Geneva Dealership 0 1 2 3 4 5 Total
0 0.0700 0.1000 0.0800 0.0300 0.0067 0.0000 0.2867
1 0.0700 0.1200 0.1100 0.0600 0.0067 0.0033 0.3700
2 0.0300 0.1400 0.0300 0.0400 0.0100 0.0067 0.2567
3 0.0100 0.0300 0.0200 0.0100 0.0167 0.0000 0.0867
Total 0.18 0.39 0.24 0.14 0.04 0.01 1.0000

© 2024 Cengage Group. All Rights Reserved.

 5.4 Probability Distribution for Total Daily Sales at
DiCarlo Motors
Let us define the total sales at DiCarlo Motors as: .
From the bivariate probability distribution, we see that: , , , and so on.

The table to the right shows the s f(s) s f(s) s − E(s) [s − E(s)]2 [s − E(s)]2 f(s)
complete probability distribution, 0 0.0700 0.0000 -2.6433 6.9872 0.4891
along with the computation of 1 0.1700 0.1700 -1.6433 2.7005 0.4591
2 0.2300 0.4600 -0.6433 0.4139 0.0952
the expected value and 3 0.2900 0.8700 0.3567 0.1272 0.0369
variance of total daily sales at 4 0.1267 0.5067 1.3567 1.8405 0.2331
DiCarlo Motors: 5 0.0667 0.3333 2.3567 5.5539 0.3703
6 0.0233 0.1400 3.3567 11.2672 0.2629
7 0.0233 0.1633 4.3567 18.9805 0.4429
8 0.0000 0.0000 5.3567 28.6939 0.0000
E(s) = 2.6433 Var(s) = 2.3895

© 2024 Cengage Group. All Rights Reserved.

 5.4 Expected Value and Variance of Daily Sales at the
Geneva Dealership
With bivariate probability distributions, we often want to better understand the relationship
existing between the two random variables by calculating the covariance and/or correlation
coefficient as measures of association between two random variables.
To calculate the covariance, we need , , and .
We have already computed the last two. Calculations of daily sales at the Geneva dealership
follow:
x f(x) x f(x) x − E(x) [x − E(x)]2 [x − E(x)] f(x)
2

0 0.2867 0.0000 -1.1435 1.3076 0.3748

1 0.3700 0.3700 -0.1435 0.0206 0.0076
2 0.2567 0.5134 0.8565 0.7336 0.1883
3 0.0867 0.2601 1.8565 3.4466 0.2988
E(x) = 1.1435 Var(x) = 0.8696

© 2024 Cengage Group. All Rights Reserved.

 5.4 Covariance and Correlation at DiCarlo Motors

The formula to compute the covariance between the random variables and is

The covariance of 0.1350 indicates that daily sales at DiCarlo’s two dealerships have a pos­itive
relationship, but it is not suitable to get a sense of the strength of such relationship.
To assess the strength of a relationship, we use the correlation coefficient between x and y

A correlation coefficient of .1295 indicates that there is a weak positive relationship between
the random variables representing daily sales at the two DiCarlo dealerships.
If the correlation coefficient had equaled zero, we would have concluded that daily sales at the
two dealerships were independent.

© 2024 Cengage Group. All Rights Reserved.

 5.4 Financial Applications

A financial advisor is considering a financial portfolio for the coming year and for each of the
four economic scenarios has developed a probability distribu­tion showing:
x = the percent return for investing in a large-­cap stock fund the percent return
y = the percent return for investing in a long-term government bond fund
The bivariate probability distribution for x and y is simply a list with a separate row for each
experimental outcome (economic scenario.)
Large-Cap Long-Term
Economic Probability Stock Fund Government Bond
Scenario f(x, y) (x) Fund (y)
Recession 0.10 −40 30
Weak Growth 0.25 5 5
Stable Growth 0.50 15 4
Strong Growth 0.15 30 2

© 2024 Cengage Group. All Rights Reserved.

 5.4 Return and Risk of the Individual Funds

We can compute the expected percent return for investing in the stock fund, E(x), and the
expected percent return for investing in the bond fund, E(y).

Financial analysts recommend that investors also consider the standard deviation associated
with an investment as a measure of risk.
First, we need to calculate the individual variances.

Thus, the individual standard deviation for x and y are:

and

© 2024 Cengage Group. All Rights Reserved.

5.4 A Financial Portfolio of a Stock Fund and a Bond Fund

The computations of expected value and standard deviation for each of the two funds reveal
that the large-cup stock fund x offers a more attractive return:

Conversely, investing in the long-term bond fund y is less risky:

The choice of investment may depend on our attitude toward risk and return:
• An aggressive investor may choose the stock fund because of the higher return.
• A conservative investor may choose the bond fund because of the lower risk.
Another option is investing in a portfolio consisting of a linear combination of the stock fund and
the bond fund.

© 2024 Cengage Group. All Rights Reserved.

 5.4 Properties of the Linear Combination of x and y

Let us define a random variable return, , as a linear combination of x and y:

Where a and b are coefficients such that and .
To build a financial portfolio made of 50% stock fund and 50% bond fund, we calculate as

Thus, a $100 investment would return $7.90.

The variance of a linear combination of two random variables and can be calculated as

Where is the covariance of and .

Given , variance and standard deviation of the financial portfolio are

© 2024 Cengage Group. All Rights Reserved.

5.4 Risk and Return for the Three Financial Alternatives

The expected return of the financial portfolio consisting of investing 50% in the stock fund and
50% in the bond fund is halfway between that of the stock fund alone and the bond fund
alone, but at a considerably lower risk, as shown by the smaller standard deviation.
Other linear combinations are possible. For example, a portfolio consisting of 25% stock fund
and 75% bond fund reveals a comparable return to the 50%-50% investment, but at an even
lower risk:
Expected Variance Standard Deviation
Investment Alternative Return (%) of Return of Return (%)
100% in Stock Fund 9.25 328.1875 18.1159
100% in Bond Fund 6.55 61.9475 7.8707
Portfolio (50% in stock 7.90 29.8650 5.4650
fund, 50% in bond fund)

© 2024 Cengage Group. All Rights Reserved.

 5.5 The Binomial Experiment

A binomial experiment exhibits the following four properties.

1. The experiment consists of a sequence of n identical trials.
2. Two outcomes, success and failure, are possible in each trial.
3. The probability of a success, denoted by p, and the probability of a failure, denoted by
1 – p, do not change from trial to trial (the stationarity assumption. *See notes.)
4. The trials are independent.
If properties 2-4 are present, we say the trials are generated by a Bernoulli process.
In a binomial experiment, our interest is in the number x of successes occurring in the n trials.
Because the number of successes is finite, x is a discrete random variable.
The probability distribution associated with this random variable is called the binomial
probability distribution, and it is described by the binomial probability function.

© 2024 Cengage Group. All Rights Reserved.

 5.5 Martin Clothing Store Problem

On the basis of past experience, the store manager at the Martin Clothing store estimates
the probability that any one customer will make a purchase to be 0.30.
What is the probability that two of the next three
customers will make a purchase?
The tree diagram to the right shows that the
experiment has eight possible outcomes.
The requirements for a binomial experiment are met:
1. Three identical trials, one for each customer
entering the store
2. Two outcomes: S = purchase, F = no purchase
3. Purchase: p = 0.30; no purchase: 1 – p = 0.70
4. Each purchase decision is independent

© 2024 Cengage Group. All Rights Reserved.

 5.5 The Binomial Probability Function
The binomial probability function determines the probability of observing successes in trials
in a binomial experiment.

Where:
= the number of successes
= the probability of a success in one trial

𝑓(𝑥)
= the number of trials
= the probability of x successes in n trials

Table 5 of Appendix B gives the probability of successes in trials for a bi­nomial experiment
with varying values of the probability of success, p (*see notes.)

© 2024 Cengage Group. All Rights Reserved.

 5.5 The Binomial Probability Calculation for the Martin
Clothing Store Problem
The binomial probability function is the product of two factors:
1. The number of experimental outcomes providing exactly successes in trials:

2. The probability of a particular sequence of trial outcomes with successes in trials:

Thus, the probability that two () of the next three () customers will make a purchase at the
Martin Clothing Store is

© 2024 Cengage Group. All Rights Reserved.

 5.5 The Binomial Probability Distribution for the
Martin Clothing Store Problem
If we repeat the calculation of the previous slide for the probability that the number of
customers making a purchase is either 0, 1, or 3, we obtain the probability distribution.
The probability distribution may also be displayed with a graph, in which the height of a bar is
proportional to the probability for a given value of .
x f(x)
0

© 2024 Cengage Group. All Rights Reserved.

 5.5 Properties of the Binomial Distribution
In the previous section, we provided formulas for computing the expected value and variance of
a discrete random variable.
If the random variable has a bi­nomial distribution with a known number of trials and a known
probability of success, the general formulas for the expected value and variance are

For the Martin Clothing store problem, we have:

customers
customers
If the next month the Martin Clothing store forecasts 1,000 customers, we can expect
customers making a purchase
with a standard deviation of customers

© 2024 Cengage Group. All Rights Reserved.

 5.6 Poisson Probability Distribution
A Poisson Probability Distribution describes x, the number of occurrences over a specified
interval of time or space, if the following two properties of a Poisson experiment are
respected.
1. The probability of an occurrence is the same for any two intervals of equal length.
2. The occurrence or nonoccurrence in any interval is independent of the occur­rence or
nonoccurrence in any other interval.
The Poisson probability function is described by the following equation

Where:
x= a discrete random variable describing the number of occurrences over an
interval.
f(x) = the probability of x occurrences in an interval
m = expected value or mean number of occurrences in an interval
Also, in a Poisson probability distribution, (the variance is equal to the mean.)

© 2024 Cengage Group. All Rights Reserved.

5.6 An Application of the Poisson Probability Distribution
An analysis of historical data shows that patients arrive at the emergency room of a large
hospital on weekend mornings at the average rate over a 15-minute interval.
What is the probability of exactly arrivals in 15 minutes on a weekend morning?
In a Poisson experiment, we have

What if we want to know the probability of exactly arrival over a 3-minute interval?
We observe that, on average, over a 3-minute interval, we have . Thus, the probability of one
arrival over a 3-minute period is

The probability of observing 5 arrivals over a 15-minute period is not the same as observing 1
arrival over a 3-minute period.

© 2024 Cengage Group. All Rights Reserved.

 5.6 Properties of the Poisson Probability Distribution
Because there is no stated upper limit for the number of occurrences, the Poisson probability
function is applicable for values without limit.
In practical applications, will eventually become large enough so that becomes approximately
zero and the probability of any larger values of negligible.
A Poisson probability distribution has the variance equal to the mean:

Probabilities can be calculated using a Poisson probability table (Table 7, Appendix B of the
textbook).
For example, to obtain the probability of arrivals over a 15-minute period when using the table,
follow these two steps:
1. Scroll down the leftmost column until you find
2. Move along the row for until you reach the column corresponding to .
The answer is 0.0378. As shown before, if using and , we get 0.2707 (*see notes.)

© 2024 Cengage Group. All Rights Reserved.

 5.7 Hypergeometric Probability Distribution
The hypergeometric probability distribution is closely related to the binomial distribution, but
it differs in two key respects:
• The trials are not independent.
• The probability of success changes from trial to trial.
The hypergeometric probability function is described by the following equation:

Where: x = the number of successes

𝑓(𝑥) =
n= the number of trials
the probability of x successes in n trials
N= the number of elements in the population
r= the number of elements in the population labeled success

© 2024 Cengage Group. All Rights Reserved.

 5.7 Hypergeometric Probability Function
The hypergeometric probability function, introduced in the previous slide, is used to compute
the probability that in a random selection of n elements, selected without replacement, we
obtain x elements labeled success and n − x elements labeled failure.
For this outcome to occur, we must obtain x successes from the r successes in the population
and n − x failures from the N − r failures.
The hypergeometric probability function, f(x), the probability of obtaining x successes in n trials,
shown in the previous slide, is made of the following three components:

© 2024 Cengage Group. All Rights Reserved.

 5.7 An Application of the Hypergeometric Distribution
If a box of electric fuses contains exactly defective fuses, what is the probability that a quality
inspector will find exactly defective fuse out of selected?

If we want instead the probability of finding at least one defective fuse (), it is easier to find the
probability that there are no defective fuses (), and then take the complement.

Thus, the probability of finding at least one defective fuse is

© 2024 Cengage Group. All Rights Reserved.

 5.7 Properties of the Hypergeometric Distribution
The expected value (mean) of a hyper-geometric distribution is calculated as

In the electric fuses example, we have , , and . Thus

The variance of a hypergeometric distribution can be calculated as

In the electric fuses example, we can calculate the standard deviation as

© 2024 Cengage Group. All Rights Reserved.

 Summary
• A random variable provides a numerical description of the outcome of an experiment.
• For any discrete random variable, the probability distribution defined by a probability
function provides the probability associated with each value of the random variable.
• We introduced two types of discrete probability distributions:
• A table consisting of the values of the random variable and associated probabilities.
• A mathematical function that provides the probabilities for the random variable.
• We described a probability distribution with the expected value, a measure of central
location, and the variance and standard deviation as measures of variability.
• We showed how to compute the covariance and correlation coefficient as measures of a
bivariate relationship, and how bivariate distributions involving market returns on financial
assets can be used to create financial portfolios.
• Finally, we introduced three discrete probability distributions that are described by a
mathematical function: the binomial, Poisson, and hypergeometric distributions.

© 2024 Cengage Group. All Rights Reserved.