1

1
1
1
1
52 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

2
1
PROBABILITY

2–1
2–2
Using Statistics 51
Basic Definitions: Events, Sample Space,
and Probabilities 53
2–3 Basic Rules for Probability 57
2–4 Conditional Probability 61
2–5 Independence of Events 66
2–6 Combinatorial Concepts 70

1
2–7 The Law of Total Probability and Bayes’ Theorem 73
2–8 The Joint Probability Table 79
2–9 Using the Computer 80
2–10 Summary and Review of Terms 84
Case 2 Job Applications 89

1 LEARNING OBJECTIVES

After studying this chapter, you should be able to:

1
• Define probability, sample space, and event.
• Distinguish between subjective and objective probability.
• Describe the complement of an event and the intersection
and union of two events.
• Compute probabilities of various types of events.

1
• Explain the concept of conditional probability and how to
compute it.
• Describe permutation and combination and their use in certain
probability computations.
• Explain Bayes’ theorem and its application.

1 50
1
1
1
1
1
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 53
Complete Business Companies, 2009
Statistics, Seventh Edition

1
2–1 Using Statistics

A Missed Pickup Is a Missed Opportunity

1
A bizarre sequence of events took place on the University of California campus
at Berkeley on October 20, 2003. An employee of the university took a package
containing 30 applications by graduate students at the university for the presti-
gious Fulbright Fellowship, administered by the U.S. Department of Education,
and dropped them at the Federal Express pickup box on campus. October 20 was
the deadline the Department of Education had set for posting by each university

1
of all its applications for awards on behalf of its students.
But just that day, something that had never happened before took place.
Because of a “computer glitch,” as Federal Express later described it, there was no
pickup by the company from its box on Sproul Plaza on the U.C. campus. When
the problem became apparent to the university, an employee sent an e-mail mes-
sage late that night to the Department of Education in Washington, apologizing
for the mishap, which was not the University’s fault, and requesting an extension
of time for its students. The Department of Education refused.

1
There ensued a long sequence of telephone calls, and the Chancellor of
the University, Robert M. Berdahl, flew to Washington to beg the authorities
to allow his students to be considered. The Department of Education refused.
At one point, one of the attorneys for the Department told the University that
had the e-mail message not been sent, everything would have been fine since
FedEx would have shown the date of posting as October 20. But since the
e-mail message had been sent, the fate of the applications was sealed. Usually,

1
15 out of 30 applications from U.C. Berkeley result in awards. But because of
this unfortunate sequence of events, no Berkeley graduate students were to
receive a Fulbright Fellowship in 2004.

Dean E. Murphy, “Missed Pickup Means a Missed Opportunity for 30 Seeking a Fellowship,” The New York
Times, February 5, 2004, p. A14.

This story demonstrates how probabilities affect everything in our lives. A priori,
there was an extremely small chance that a pickup would be missed: According to
FedEx this simply doesn’t happen. The university had relied on the virtually sure
probability of a pickup, and thus posted the applications on the last possible day.
Moreover, the chance that an employee of the university would find out that the pick-
up was missed on that same day and e-mail the Department of Education was very
small. Yet the sequence of rare events took place, with disastrous results for the grad-
uate students who had worked hard to apply for these important awards.
A probability is a quantitative measure of uncertainty—a number that conveys
the strength of our belief in the occurrence of an uncertain event. Since life is full of
uncertainty, people have always been interested in evaluating probabilities. The stat-
istician I. J. Good suggests that “the theory of probability is much older than the
human species,” since the assessment of uncertainty incorporates the idea of learning
from experience, which most creatures do.1
1
I. J. Good, “Kinds of Probability,” Science, no. 129 (February 20, 1959), pp. 443–47.
54 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

52 Chapter 2

The theory of probability as we know it today was largely developed by European

mathematicians such as Galileo Galilei (1564–1642), Blaise Pascal (1623–1662), Pierre
de Fermat (1601–1665), Abraham de Moivre (1667–1754), and others.
As in India, the development of probability theory in Europe is often associated
with gamblers, who pursued their interests in the famous European casinos, such as
the one at Monte Carlo. Many books on probability and statistics tell the story of the
Chevalier de Mère, a French gambler who enlisted the help of Pascal in an effort to
obtain the probabilities of winning at certain games of chance, leading to much of the
European development of probability.
Today, the theory of probability is an indispensable tool in the analysis of
situations involving uncertainty. It forms the basis for inferential statistics as well as
for other fields that require quantitative assessments of chance occurrences, such as
quality control, management decision analysis, and areas in physics, biology, engi-
neering, and economics.
While most analyses using the theory of probability have nothing to do with
games of chance, gambling models provide the clearest examples of probability and
its assessment. The reason is that games of chance usually involve dice, cards, or
roulette wheels—mechanical devices. If we assume there is no cheating, these
mechanical devices tend to produce sets of outcomes that are equally likely, and this
allows us to compute probabilities of winning at these games.
Suppose that a single die is rolled and that you win a dollar if the number 1 or 2
appears. What are your chances of winning a dollar? Since there are six equally likely
numbers (assuming the die is fair) and you win as a result of either of two numbers
appearing, the probability that you win is 2
6, or 1
3.
As another example, consider the following situation. An analyst follows the
price movements of IBM stock for a time and wants to assess the probability that
the stock will go up in price in the next week. This is a different type of situation.
The analyst does not have the luxury of a known set of equally likely outcomes,
where “IBM stock goes up next week” is one of a given number of these equally
likely possibilities. Therefore, the analyst’s assessment of the probability of the event
will be a subjective one. The analyst will base her or his assessment of this probability
on knowledge of the situation, guesses, or intuition. Different people may assign dif-
ferent probabilities to this event depending on their experience and knowledge,
hence the name subjective probability.
Objective probability is probability based on symmetry of games of chance or
similar situations. It is also called classical probability. This probability is based on the
idea that certain occurrences are equally likely (the term equally likely is intuitively
clear and will be used as a starting point for our definitions): The numbers 1, 2, 3, 4, 5,
and 6 on a fair die are each equally likely to occur. Another type of objective prob-
ability is long-term relative-frequency probability. If, in the long run, 20 out of 1,000 con-
sumers given a taste test for a new soup like the taste, then we say that the probability
that a given consumer will like the soup is 20
1,000
0.02. If the probability that a
head will appear on any one toss of a coin is 1
2, then if the coin is tossed a large num-
ber of times, the proportion of heads will approach 1
2. Like the probability in games
of chance and other symmetric situations, relative-frequency probability is objective
in the sense that no personal judgment is involved.
Subjective probability, on the other hand, involves personal judgment, infor-
mation, intuition, and other subjective evaluation criteria. The area of subjective
probability—which is relatively new, having been first developed in the 1930s—is
somewhat controversial.2 A physician assessing the probability of a patient’s recovery
and an expert assessing the probability of success of a merger offer are both making a
personal judgment based on what they know and feel about the situation. Subjective

2
The earliest published works on subjective probability are Frank Ramsey’s The Foundation of Mathematics and Other
Logical Essays (London: Kegan Paul, 1931) and the Italian statistician Bruno de Finetti’s “La Prévision: Ses Lois Logiques,
Ses Sources Subjectives,” Annales de L’Institut Henri Poincaré 7, no. 1 (1937).
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 55
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 53

probability is also called personal probability. One person’s subjective probability may
very well be different from another person’s subjective probability of the same event.
Whatever the kind of probability involved, the same set of mathematical rules
holds for manipulating and analyzing probability. We now give the general rules for
probability as well as formal definitions. Some of our definitions will involve counting
the number of ways in which some event may occur. The counting idea is imple-
mentable only in the case of objective probability, although conceptually this idea
may apply to subjective probability as well, if we can imagine a kind of lottery with a
known probability of occurrence for the event of interest.

2–2 Basic Definitions: Events, Sample Space,

and Probabilities
To understand probability, some familiarity with sets and with operations involving
sets is useful.

A set is a collection of elements.

The elements of a set may be people, horses, desks, cars, files in a cabinet, or even
numbers. We may define our set as the collection of all horses in a given pasture, all
people in a room, all cars in a given parking lot at a given time, all the numbers
between 0 and 1, or all integers. The number of elements in a set may be infinite, as
in the last two examples.
A set may also have no elements.

The empty set is the set containing no elements. It is denoted by .

We now define the universal set.

The universal set is the set containing everything in a given context. We

denote the universal set by S.

Given a set A, we may define its complement.

The complement of set A is the set containing all the elements in the uni-
versal set S that are not members of set A. We denote the complement of
A by A. The set A is often called “not A.”

A Venn diagram is a schematic drawing of sets that demonstrates the relationships

between different sets. In a Venn diagram, sets are shown as circles, or other closed
figures, within a rectangle corresponding to the universal set, S. Figure 2–1 is a Venn
diagram demonstrating the relationship between a set A and its complement A.
As an example of a set and its complement, consider the following. Let the uni-
versal set S be the set of all students at a given university. Define A as the set of all
students who own a car (at least one car). The complement of A, or A, is thus the set
of all students at the university who do not own a car.
Sets may be related in a number of ways. Consider two sets A and B within the
context of the same universal set S. (We say that A and B are subsets of the universal
set S.) If A and B have some elements in common, we say they intersect.

The intersection of A and B, denoted A
B, is the set containing all ele-

ments that are members of both A and B.

When we want to consider all the elements of two sets A and B, we look at their
union.

The union of A and B, denoted A  B, is the set containing all elements

that are members of either A or B or both.
56 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

54 Chapter 2

FIGURE 2–1 A Set A and Its Complement A FIGURE 2–2 Sets A and B and Their Intersection

U
A A B B
A

FIGURE 2–3 The Union of A and B FIGURE 2–4 Two Disjoint Sets

AUB A B

As you can see from these definitions, the union of two sets contains the intersec-
tion of the two sets. Figure 2–2 is a Venn diagram showing two sets A and B and
their intersection A
B. Figure 2–3 is a Venn diagram showing the union of the
same sets.
As an example of the union and intersection of sets, consider again the set of all
students at a university who own a car. This is set A. Now define set B as the set of
all students at the university who own a bicycle. The universal set S is, as before, the
set of all students at the university. And A
B is the intersection of A and B—it is the
set of all students at the university who own both a car and a bicycle. And A  B is
the union of A and B—it is the set of all students at the university who own either a
car or a bicycle or both.
Two sets may have no intersection: They may be disjoint. In such a case, we say
that the intersection of the two sets is the empty set . In symbols, when A and B are
disjoint, A
B
. As an example of two disjoint sets, consider the set of all
students enrolled in a business program at a particular university and all the students
at the university who are enrolled in an art program. (Assume no student is enrolled
in both programs.) A Venn diagram of two disjoint sets is shown in Figure 2–4.
In probability theory we make use of the idea of a set and of operations involving
sets. We will now provide some basic definitions of terms relevant to the computation
of probability. These are an experiment, a sample space, and an event.

An experiment is a process that leads to one of several possible outcomes.

An outcome of an experiment is some observation or measurement.

Drawing a card out of a deck of 52 cards is an experiment. One outcome of the

experiment may be that the queen of diamonds is drawn.
A single outcome of an experiment is called a basic outcome or an elementary event.
Any particular card drawn from a deck is a basic outcome.

The sample space is the universal set S pertinent to a given experiment.

The sample space is the set of all possible outcomes of an experiment.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 57
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 55

FIGURE 2–5 Sample Space for Drawing a Card

The event A, “an A A A A The outcome “ace of spades”

ace is drawn.” means that event A has
K K K K
occurred.
Q Q Q Q
J J J J
10 10 10 10
9 9 9 9
8 8 8 8
7 7 7 7
6 6 6 6
5 5 5 5
4 4 4 4
3 3 3 3
2 2 2 2

The sample space for the experiment of drawing a card out of a deck is the set of all
cards in the deck. The sample space for an experiment of reading the temperature is
the set of all numbers in the range of temperatures.

An event is a subset of a sample space. It is a set of basic outcomes. We say

that the event occurs if the experiment gives rise to a basic outcome
belonging to the event.

For example, the event “an ace is drawn out of a deck of cards” is the set of the four
aces within the sample space consisting of all 52 cards. This event occurs whenever
one of the four aces (the basic outcomes) is drawn.
The sample space for the experiment of drawing a card out of a deck of 52 cards is
shown in Figure 2–5. The figure also shows event A, the event that an ace is drawn.
In this context, for a given experiment we have a sample space with equally likely
basic outcomes. When a card is drawn out of a well-shuffled deck, every one of the cards
(the basic outcomes) is as likely to occur as any other. In such situations, it seems rea-
sonable to define the probability of an event as the relative size of the event with respect
to the size of the sample space. Since a deck has 4 aces and 52 cards, the size of A is 4
and the size of the sample space is 52. Therefore, the probability of A is equal to 4
52.
The rule we use in computing probabilities, assuming equal likelihood of all basic
outcomes, is as follows:

Probability of event A:
n (A)
P(A) = (2–1)
n (S)
where
n(A)
the number of elements in the set of the event A
n(S)
the number of elements in the sample space S
58 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

56 Chapter 2

FIGURE 2–6 The Events A and ♥ and Their Union and Intersection

Union of A and A A A A
(everything that K K K K
is circled at least once).
Q Q Q Q
Event A J J J J
10 10 10 10
9 9 9 9
8 8 8 8
Event 7 7 7 7
6 6 6 6
The intersection of A
and comprises
5 5 5 5 the points circled
4 4 4 4 twice: the ace
of hearts.
3 3 3 3
2 2 2 2

The probability of drawing an ace is P (A)
n(A)
n(S)
4
52.

EXAMPLE 2–1 Roulette is a popular casino game. As the game is played in Las Vegas or Atlantic
City, the roulette wheel has 36 numbers, 1 through 36, and the number 0 as well as
the number 00 (double zero). What is the probability of winning on a single number
that you bet?

Solution The sample space S in this example consists of 38 numbers (0, 00, 1, 2, 3,…, 36),
each of which is equally likely to come up. Using our counting rule P (any one given
number)
1
38.

Let’s now demonstrate the meaning of union and intersection with the example
of drawing a card from a deck. Let A be the event that an ace is drawn and ♥ the
event that a heart is drawn. The sample space is shown in Figure 2–6. Note that
the event A
♥ is the event that the card drawn is both an ace and a heart (i.e., the
ace of hearts). The event A  ♥ is the event that the card drawn is either an ace or a
heart or both.

PROBLEMS

2–1. What are the two main types of probability?

2–2. What is an event? What is the union of two events? What is the intersection of
two events?
2–3. Define a sample space.
2–4. Define the probability of an event.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 59
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 57

2–5. Let G be the event that a girl is born. Let F be the event that a baby over
5 pounds is born. Characterize the union and the intersection of the two events.
2–6. Consider the event that a player scores a point in a game against team A and
the event that the same player scores a point in a game against team B. What is the
union of the two events? What is the intersection of the two events?
2–7. A die is tossed twice and the two outcomes are noted. Draw the Venn diagram
of the sample space and indicate the event “the second toss is greater than the first.”
Calculate the probability of the event.
2–8. Ford Motor Company advertises its cars on radio and on television. The com-
pany is interested in assessing the probability that a randomly chosen person is
exposed to at least one of these two modes of advertising. If we define event R as the
event that a randomly chosen person was exposed to a radio advertisement and
event T as the event that the person was exposed to a television commercial, define
R  T and R
T in this context.
2–9. A brokerage firm deals in stocks and bonds. An analyst for the firm is inter-
ested in assessing the probability that a person who inquires about the firm will even-
tually purchase stock (event S) or bonds (event B). Define the union and the
intersection of these two events.
2–10. The European version of roulette is different from the U.S. version in that the
European roulette wheel doesn’t have 00. How does this change the probability of win-
ning when you bet on a single number? European casinos charge a small admission
fee, which is not the case in U.S. casinos. Does this make sense to you, based on your
answer to the earlier question?

2–3 Basic Rules for Probability

We have explored probability on a somewhat intuitive level and have seen rules that
help us evaluate probabilities in special cases when we have a known sample space
with equally likely basic outcomes. We will now look at some general probability
rules that hold regardless of the particular situation or kind of probability (objective
or subjective). First, let us give a general definition of probability.

Probability is a measure of uncertainty. The probability of event A is a

numerical measure of the likelihood of the event’s occurring.

The Range of Values

Probability obeys certain rules. The first rule sets the range of values that the proba-
bility measure may take.

For any event A, the probability P(A) satisfies

0 P(A) 1 (2–2)

When an event cannot occur, its probability is zero. The probability of the empty set
is zero: P()
0. In a deck where half the cards are red and half are black, the prob-
ability of drawing a green card is zero because the set corresponding to that event is
the empty set: There are no green cards.
Events that are certain to occur have probability 1.00. The probability of the
entire sample space S is equal to 1.00: P(S)
1.00. If we draw a card out of a deck,
1 of the 52 cards in the deck will certainly be drawn, and so the probability of the
sample space, the set of all 52 cards, is equal to 1.00.
60 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

58 Chapter 2

FIGURE 2–7 Interpretation of a Probability

0 0.25 0.5 0.75 1

Event is not very Event is as Event is very

likely to occur. likely to occur likely to occur.
as not to occur.
Event is more
likely not to Event is more
occur than likely to occur
to occur. than not to occur.

Within the range of values 0 to 1, the greater the probability, the more confidence
we have in the occurrence of the event in question. A probability of 0.95 implies a
very high confidence in the occurrence of the event. A probability of 0.80 implies
a high confidence. When the probability is 0.5, the event is as likely to occur as it is
not to occur. When the probability is 0.2, the event is not very likely to occur. When
we assign a probability of 0.05, we believe the event is unlikely to occur, and so on.
Figure 2–7 is an informal aid in interpreting probability.
Note that probability is a measure that goes from 0 to 1. In everyday conversation
we often describe probability in less formal terms. For example, people sometimes
talk about odds. If the odds are 1 to 1, the probability is 1
2; if the odds are 1 to 2,
the probability is 1
3; and so on. Also, people sometimes say, “The probability is 80
percent.” Mathematically, this probability is 0.80.

The Rule of Complements

Our second rule for probability defines the probability of the complement of an
event in terms of the probability of the original event. Recall that the complement of
set A is denoted by A.

Probability of the complement:

P(A)
1  P(A) (2–3)

As a simple example, if the probability of rain tomorrow is 0.3, then the probability of
no rain tomorrow must be 1  0.3
0.7. If the probability of drawing an ace is 4
52,
then the probability of the drawn card’s not being an ace is 1  4
52
48
52.
The Rule of Unions. We now state a very important rule, the rule of unions. The
rule of unions allows us to write the probability of the union of two events in terms of
the probabilities of the two events and the probability of their intersection:3

The rule of unions:

P(A  B)
P(A)  P(B)  P(A
B) (2–4)

3
The rule can be extended to more than two events. In the case of three events, we have P(A  B  C)
P(A)  P(B) 
P(C)  P(A
B)  P(A
C)  P(B
C)  P(A
B
C). With more events, this becomes even more complicated.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 61
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 59

[The probability of the intersection of two events P(A
B) is called their joint
probability.] The meaning of this rule is very simple and intuitive: When we add the
probabilities of A and B, we are measuring, or counting, the probability of their inter-
section twice—once when measuring the relative size of A within the sample space and
once when doing this with B. Since the relative size, or probability, of the intersection of
the two sets is counted twice, we subtract it once so that we are left with the true proba-
bility of the union of the two events (refer to Figure 2–6). Note that instead of finding the
probability of A  B by direct counting, we can use the rule of unions: We know that the
probability of an ace is 4
52, the probability of a heart is 13
52, and the probability of
their intersection—the drawn card being the ace of hearts—is 1
52. Thus, P(A  ♥)

4
52  13
52  1
52
16
52, which is exactly what we find from direct counting.
The rule of unions is especially useful when we do not have the sample space for
the union of events but do have the separate probabilities. For example, suppose
your chance of being offered a certain job is 0.4, your probability of getting another
job is 0.5, and your probability of being offered both jobs (i.e., the intersection) is 0.3.
By the rule of unions, your probability of being offered at least one of the two jobs
(their union) is 0.4  0.5  0.3
0.6.

Mutually Exclusive Events

When the sets corresponding to two events are disjoint (i.e., have no intersection), the
two events are called mutually exclusive (see Figure 2–4). For mutually exclusive
events, the probability of the intersection of the events is zero. This is so because
the intersection of the events is the empty set, and we know that the probability of
the empty set  is zero.

For mutually exclusive events A and B:

P(A
B)
0 (2–5)

This fact gives us a special rule for unions of mutually exclusive events. Since the
probability of the intersection of the two events is zero, there is no need to subtract
P(A
B) when the probability of the union of the two events is computed. Therefore,

For mutually exclusive events A and B:

P(A  B)
P(A)  P(B) (2–6)

This is not really a new rule since we can always use the rule of unions for the union
of two events: If the events happen to be mutually exclusive, we subtract zero as the
probability of the intersection.
To continue our cards example, what is the probability of drawing either a heart or
a club? We have P(♥  ♣)
P(♥)  P(♣)
13
52  13
52
26
52
1
2. We need
not subtract the probability of an intersection, since no card is both a club and a heart.

PROBLEMS

2–11. According to an article in Fortune, institutional investors recently changed the

proportions of their portfolios toward public sector funds.4 The article implies that 8%
of investors studied invest in public sector funds and 6% in corporate funds. Assume
that 2% invest in both kinds. If an investor is chosen at random, what is the probability
that this investor has either public or corporate funds?

4
“Fueling the Fire,” Fortune, March 5, 2007, p. 60.
62 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

60 Chapter 2

2–12. According to The New York Times, 5 million BlackBerry users found their
devices nonfunctional on April 18, 2007.5 If there were 18 million users of handheld
data devices of this kind on that day, what is the probability that a randomly chosen
user could not use a device?
2–13. In problem 2–12, assume that 3 million out of 18 million users could not use
their devices as cellphones, and that 1 million could not use their devices as a cell-
phone and for data device. What is the probability that a randomly chosen device
could not be used either for data or for voice communication?
2–14. According to a report on CNN Business News in April 1995, the probability
of being murdered (in the United States) in 1 year is 9 in 100,000. How might such a
probability have been obtained?
2–15. Assign a reasonable numerical probability to the statement “Rain is very likely
tonight.”
2–16. How likely is an event that has a 0.65 probability? Describe the probability
in words.
2–17. If a team has an 80% chance of winning a game, describe its chances in words.
2–18. ShopperTrak is a hidden electric eye designed to count the number of shoppers
entering a store. When two shoppers enter a store together, one walking in front of the
other, the following probabilities apply: There is a 0.98 probability that the first shop-
per will be detected, a 0.94 probability that the second shopper will be detected, and a
0.93 probability that both of them will be detected by the device. What is the probabil-
ity that the device will detect at least one of two shoppers entering together?
2–19. A machine produces components for use in cellular phones. At any given
time, the machine may be in one, and only one, of three states: operational, out of
control, or down. From experience with this machine, a quality control engineer
knows that the probability that the machine is out of control at any moment is 0.02,
and the probability that it is down is 0.015.
a. What is the relationship between the two events “machine is out of control”
and “machine is down”?
b. When the machine is either out of control or down, a repair person must be
called. What is the probability that a repair person must be called right now?
c. Unless the machine is down, it can be used to produce a single item. What is
the probability that the machine can be used to produce a single component
right now? What is the relationship between this event and the event
“machine is down”?
2–20. Following are age and sex data for 20 midlevel managers at a service com-
pany: 34 F, 49 M, 27 M, 63 F, 33 F, 29 F, 45 M, 46 M, 30 F, 39 M, 42 M, 30 F, 48 M,
35 F, 32 F, 37 F, 48 F, 50 M, 48 F, 61 F. A manager must be chosen at random to
serve on a companywide committee that deals with personnel problems. What is the
probability that the chosen manager will be either a woman or over 50 years old or
both? Solve both directly from the data and by using the law of unions. What is the
probability that the chosen manager will be under 30?
2–21. Suppose that 25% of the population in a given area is exposed to a television
commercial for Ford automobiles, and 34% is exposed to Ford’s radio advertisements.
Also, it is known that 10% of the population is exposed to both means of advertising.
If a person is randomly chosen out of the entire population in this area, what is the
probability that he or she was exposed to at least one of the two modes of advertising?
2–22. Suppose it is known that 85% of the people who inquire about investment
opportunities at a brokerage house end up purchasing stock, and 33% end up
purchasing bonds. It is also known that 28% of the inquirers end up getting a portfolio

5
Brad Stone, “Bereft of BlackBerrys, the Untethered Make Do,” The New York Times, April 19, 2007, p. C1.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 63
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 61

with both stocks and bonds. If a person is just making an inquiry, what is the probabil-
ity that she or he will get stock or bonds or both (i.e., open any portfolio)?
2–23. A firm has 550 employees; 380 of them have had at least some college edu-
cation, and 412 of the employees underwent a vocational training program. Further-
more, 357 employees both are college-educated and have had the vocational
training. If an employee is chosen at random, what is the probability that he or she is
college-educated or has had the training or both?
2–24. In problem 2–12, what is the probability that a randomly chosen user could
use his or her device?
2–25. As part of a student project for the 1994 Science Fair in Orange, Massachusetts,
28 horses were made to listen to Mozart and heavy-metal music. The results were as
follows: 11 of the 28 horses exhibited some head movements when Mozart was played;
8 exhibited some head movements when the heavy metal was played; and 5 moved
their heads when both were played. If a horse is chosen at random, what is the proba-
bility the horse exhibited head movements to Mozart or to heavy metal or to both?

2–4 Conditional Probability

As a measure of uncertainty, probability depends on information. Thus, the
probability you would give the event “Xerox stock price will go up tomorrow”
depends on what you know about the company and its performance; the probability
is conditional upon your information set. If you know much about the company, you
may assign a different probability to the event than if you know little about the com-
pany. We may define the probability of event A conditional upon the occurrence of
event B. In this example, event A may be the event that the stock will go up tomor-
row, and event B may be a favorable quarterly report.

The conditional probability of event A given the occurrence of event B is

P(A ¨ B)
P(A ƒ B)
(2–7)
P(B)
assuming P(B)
0.

The vertical line in P(A | B) is read given, or conditional upon. The probability of event
A given the occurrence of event B is defined as the probability of the intersection of
A and B, divided by the probability of event B.

As part of a drive to modernize the economy, the government of an eastern Euro- EXAMPLE 2–2
pean country is pushing for starting 100 new projects in computer development
and telecommunications. Two U.S. giants, IBM and AT&T, have signed contracts

The 100 Projects

Telecommunications: T

Computers: C

AT&T IBM
64 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

62 Chapter 2

for these projects: 40 projects for IBM and 60 for AT&T. Of the IBM projects,
30 are in the computer area and 10 are in telecommunications; of the AT&T proj-
ects, 40 are in telecommunications and 20 are in computer areas. Given that a ran-
domly chosen project is in telecommunications, what is the probability that it is
undertaken by IBM?

Solution
P (IBM ¨ T) 10>100
P (IBM|T) = = = 0.2
P (T) 50>100

But we see this directly from the fact that there are 50 telecommunications projects
and 10 of them are by IBM. This confirms the definition of conditional probability in
an intuitive sense.

When two events and their complements are of interest, it may be convenient to
arrange the information in a contingency table. In Example 2–2 the table would be
set up as follows:
AT&T IBM Total
Telecommunications 40 10 50
Computers 20 30 50
Total 60 40 100

Contingency tables help us visualize information and solve problems. The definition
of conditional probability (equation 2–7) takes two other useful forms.

Variation of the conditional probability formula:

P(A
B)
P (A | B)P(B)
and
P (A
B)
P(B | A)P(A) (2–8)

These are illustrated in Example 2–3.

EXAMPLE 2–3 A consulting firm is bidding for two jobs, one with each of two large multinational
corporations. The company executives estimate that the probability of obtaining
the consulting job with firm A, event A, is 0.45. The executives also feel that if the
company should get the job with firm A, then there is a 0.90 probability that firm B
will also give the company the consulting job. What are the company’s chances of
getting both jobs?

Solution We are given P(A)
0.45. We also know that P (B | A)
0.90, and we are looking for
P(A
B), which is the probability that both A and B will occur. From the equation
we have P(A
B)
P (B | A)P (A)
0.90  0.45
0.405.

EXAMPLE 2–4 Twenty-one percent of the executives in a large advertising firm are at the top salary
level. It is further known that 40% of all the executives at the firm are women. Also,
6.4% of all executives are women and are at the top salary level. Recently, a question
arose among executives at the firm as to whether there is any evidence of salary
inequity. Assuming that some statistical considerations (explained in later chapters)
are met, do the percentages reported above provide any evidence of salary inequity?
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 65
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 63

To solve this problem, we pose it in terms of probabilities and ask whether the Solution
probability that a randomly chosen executive will be at the top salary level is approx-
imately equal to the probability that the executive will be at the top salary level given
the executive is a woman. To answer, we need to compute the probability that the
executive will be at the top level given the executive is a woman. Defining T as the
event of a top salary and W as the event that an executive is a woman, we get

P (T ¨ W) 0.064
P (T ƒ W)


0.16
P (W) 0.40

Since 0.16 is smaller than 0.21, we may conclude (subject to statistical considerations)
that salary inequity does exist at the firm, because an executive is less likely to make
a top salary if she is a woman.

Example 2–4 may incline us to think about the relations among different events.
Are different events related, or are they independent of each other? In this example,
we concluded that the two events, being a woman and being at the top salary level,
are related in the sense that the event W made event T less likely. Section 2–5 quan-
tifies the relations among events and defines the concept of independence.

PROBLEMS

2–26. SBC Warburg, Deutsche Morgan Grenfell, and UBS are foreign. Given that
a security is foreign-underwritten, find the probability that it is by SBC Warburg (see
the accompanying table).6

American Dream
Largest wholesale and investment banks
Market share, 1996, %*

10

0
Merrill Lynch

Chase Manhattan
J.P. Morgan

Goldman Sachs

Morgan Stanley

CS First Boston

Salomon Brothers
Lehman Brothers

UBS

Bear Stearns

Citicorp
Deutsche Morgan
Grenfell
SBC Warburg

DLJ
NatWest

*As % of top 25 banks. Bond and equity underwriting and placement,

M&A advice, lead management of syndicated loans and medium-term
notes.

6
From “Out of Their League?” The Economist, June 21, 1997, pp. 71–72. © 1997 The Economist Newspaper Group,
Inc. Reprinted with permission. Further reproduction prohibited. www.economist.com.
66 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

64 Chapter 2

2–27. If a large competitor will buy a small firm, the firm’s stock will rise with prob-
ability 0.85. The purchase of the company has a 0.40 probability. What is the prob-
ability that the purchase will take place and the firm’s stock will rise?
2–28. A financial analyst believes that if interest rates decrease in a given period,
then the probability that the stock market will go up is 0.80. The analyst further
believes that interest rates have a 0.40 chance of decreasing during the period in
question. Given the above information, what is the probability that the market will go
up and interest rates will go down during the period in question?
2–29. A bank loan officer knows that 12% of the bank’s mortgage holders lose their
jobs and default on the loan in the course of 5 years. She also knows that 20% of the
bank’s mortgage holders lose their jobs during this period. Given that one of her
mortgage holders just lost his job, what is the probability that he will now default on
the loan?
2–30. An express delivery service promises overnight delivery of all packages
checked in before 5 P.M. The delivery service is not perfect, however, and sometimes
delays do occur. Management knows that if delays occur in the evening flight to a
major city from which distribution is made, then a package will not arrive on time
with probability 0.25. It is also known that 10% of the evening flights to the major city
are delayed. What percentage of the packages arrive late? (Assume that all packages
are sent out on the evening flight to the major city and that all packages arrive on
time if the evening flight is not delayed.)
2–31. The following table gives numbers of claims at a large insurance company by
kind and by geographic region.

East South Midwest West

Hospitalization 75 128 29 52
Physician’s visit 233 514 104 251
Outpatient treatment 100 326 65 99

Compute column totals and row totals. What do they mean?

a. If a bill is chosen at random, what is the probability that it is from the Midwest?
b. What is the probability that a randomly chosen bill is from the East?
c. What is the probability that a randomly chosen bill is either from the Midwest
or from the South? What is the relation between these two events?
d. What is the probability that a randomly chosen bill is for hospitalization?
e. Given that a bill is for hospitalization, what is the probability that it is from
the South?
f. Given that a bill is from the East, what is the probability that it is for a
physician’s visit?
g. Given that a bill is for outpatient treatment, what is the probability that it is
from the West?
h. What is the probability that a randomly chosen bill is either from the East or
for outpatient treatment (or both)?
i. What is the probability that a randomly selected bill is either for hospitaliza-
tion or from the South (or both)?
2–32. One of the greatest problems in marketing research and other survey fields is
the problem of nonresponse to surveys. In home interviews the problem arises when
the respondent is not home at the time of the visit or, sometimes, simply refuses to
answer questions. A market researcher believes that a respondent will answer all ques-
tions with probability 0.94 if found at home. He further believes that the probability
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 67
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 65

that a given person will be found at home is 0.65. Given this information, what
percentage of the interviews will be successfully completed?
2–33. An investment analyst collects data on stocks and notes whether or not
dividends were paid and whether or not the stocks increased in price over a given period.
Data are presented in the following table.

Price No Price
Increase Increase Total
Dividends paid 34 78 112
No dividends paid 85 49 134
Total 119 127 246

a. If a stock is selected at random out of the analyst’s list of 246 stocks, what is
the probability that it increased in price?
b. If a stock is selected at random, what is the probability that it paid dividends?
c. If a stock is randomly selected, what is the probability that it both increased
in price and paid dividends?
d. What is the probability that a randomly selected stock neither paid divi-
dends nor increased in price?
e. Given that a stock increased in price, what is the probability that it also paid
dividends?
f. If a stock is known not to have paid dividends, what is the probability that it
increased in price?
g. What is the probability that a randomly selected stock was worth holding
during the period in question; that is, what is the probability that it increased
in price or paid dividends or did both?
2–34. The following table lists the number of firms where the top executive officer
made over $1 million per year. The table also lists firms according to whether share-
holder return was positive during the period in question.

Top Executive Top Executive

Made More Made Less
than $1 Million than $1 Million Total
Shareholders made money 1 6 7
Shareholders lost money 2 1 3
Total 3 7 10

a. If a firm is randomly chosen from the list of 10 firms studied, what is the
probability that its top executive made over $1 million per year?
b. If a firm is randomly chosen from the list, what is the probability that its
shareholders lost money during the period studied?
c. Given that one of the firms in this group had negative shareholder return,
what is the probability that its top executive made over $1 million?
d. Given that a firm’s top executive made over $1 million, what is the prob-
ability that the firm’s shareholder return was positive?
2–35. According to Fortune, 90% of endangered species depend on forests for the
habitat they provide.7 If 30% of endangered species are in critical danger and depend
on forests for their habitat, what is the probability that an endangered species that
depends on forests is in critical danger?

7
“Environmental Steward,” Fortune, March 5, 2007, p. 54.
68 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

66 Chapter 2

2–5 Independence of Events

In Example 2–4 we concluded that the probability that an executive made a top
salary was lower when the executive was a woman, and we concluded that the two
events T and W were not independent. We now give a formal definition of statistical
independence of events.
Two events A and B are said to be independent of each other if and only if the
following three conditions hold:

Conditions for the independence of two events A and B:

P(A | B)
P(A)
P(B | A)
P(B) (2–9)
and, most useful:
P(A
B)
P(A)P(B) (2–10)

The first two equations have a clear, intuitive appeal. The top equation says that
when A and B are independent of each other, then the probability of A stays the
same even when we know that B has occurred—it is a simple way of saying that
knowledge of B tells us nothing about A when the two events are independent.
Similarly, when A and B are independent, then knowledge that A has occurred
gives us absolutely no information about B and its likelihood of occurring.
The third equation, however, is the most useful in applications. It tells us that
when A and B are independent (and only when they are independent), we can
obtain the probability of the joint occurrence of A and B (i.e., the probability of their
intersection) simply by multiplying the two separate probabilities. This rule is thus
called the product rule for independent events. (The rule is easily derived from the
first rule, using the definition of conditional probability.)
As an example of independent events, consider the following: Suppose I roll a
single die. What is the probability that the number 6 will turn up? The answer is 1
6.
Now suppose that I told you that I just tossed a coin and it turned up heads. What is
now the probability that the die will show the number 6? The answer is unchanged,
1
6, because events of the die and the coin are independent of each other. We see
that P(6 | H)
P (6), which is the first rule above.
In Example 2–2, we found that the probability that a project belongs to IBM
given that it is in telecommunications is 0.2. We also knew that the probability that a
project belongs to IBM was 0.4. Since these two numbers are not equal, the two
events IBM and telecommunications are not independent.
When two events are not independent, neither are their complements. Therefore,
AT&T and computers are not independent events (and neither are the other two
possibilities).

EXAMPLE 2–5 The probability that a consumer will be exposed to an advertisement for a certain
product by seeing a commercial on television is 0.04. The probability that the con-
sumer will be exposed to the product by seeing an advertisement on a billboard is
0.06. The two events, being exposed to the commercial and being exposed to the
billboard ad, are assumed to be independent. (a) What is the probability that the
consumer will be exposed to both advertisements? (b) What is the probability that
he or she will be exposed to at least one of the ads?
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 69
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 67

(a) Since the two events are independent, the probability of the intersection of the two Solution
(i.e., being exposed to both ads) is P(A
B)
P(A)P(B)
0.04  0.06
0.0024.
(b) We note that being exposed to at least one of the advertisements is, by definition, the
union of the two events, and so the rule for union applies. The probability of the inter-
section was computed above, and we have P(A  B)
P(A)  P(B)  P(A
B)

0.04  0.06  0.0024
0.0976. The computation of such probabilities is important in
advertising research. Probabilities are meaningful also as proportions of the population
exposed to different modes of advertising, and are thus important in the evaluation of
advertising efforts.

Product Rules for Independent Events

The rules for the union and the intersection of two independent events extend nicely
to sequences of more than two events. These rules are very useful in random
sampling.
Much of statistics involves random sampling from some population. When we
sample randomly from a large population, or when we sample randomly with
replacement from a population of any size, the elements are independent of one
another. For example, suppose that we have an urn containing 10 balls, 3 of them red
and the rest blue. We randomly sample one ball, note that it is red, and return it to
the urn (this is sampling with replacement). What is the probability that a second ball
we choose at random will be red? The answer is still 3
10 because the second draw-
ing does not “remember” that the first ball was red. Sampling with replacement in
this way ensures independence of the elements. The same holds for random sam-
pling without replacement (i.e., without returning each element to the population
before the next draw) if the population is relatively large in comparison with the size
of the sample. Unless otherwise specified, we will assume random sampling from a
large population.
Random sampling from a large population implies independence.

Intersection Rule
The probability of the intersection of several independent events is just the
product of the separate probabilities.
The rate of defects in corks of wine bottles is very high, 75%. Assuming
independence, if four bottles are opened, what is the probability that all
four corks are defective? Using this rule: P (all 4 are defective)
P (first cork
is defective)  P (second cork is defective)  P (third cork is defective) 
P (fourth cork is defective)
0.75  0.75  0.75  0.75
0.316.
If these four bottles were randomly selected, then we would not have
to specify independence—a random sample always implies independence.

Union Rule
The probability of the union of several independent events—A1, A2,…, An—
is given by the following equation:

P(A1 ª A2 ª . . . ª An)
1  P(A1) P(A2) . . . P (An) (2–11)

The union of several events is the event that at least one of the events happens. In the
example of the wine corks, suppose we want to find the probability that at least one of
the four corks is defective. We compute this probability as follows: P (at least one is
defective)
1  P (none are defective)
1  0.25  0.25  0.25  0.25
0.99609.
70 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

68 Chapter 2

EXAMPLE 2–6 Read the accompanying article. Three women (assumed a random sample) in a
developing country are pregnant. What is the probability that at least one will die?

Poor Nations’ Mothers at Serious

Health Risk
In the industrialized world, a woman’s 7 million newborns die within a week
odds of dying from problems related to of birth because of maternal health
pregnancy are 1 in 1,687. But in the de- problems. The bank and the United
veloping world the figure is 1 in 51. The Nations are in the midst of an initiative
World Bank also says that each year to cut maternal illnesses and deaths.

Edward Epstein, “Poor Nations’ Mothers at Serious Health Risk,” World Insider, San Francisco Chronicle,
August 10, 1993, p. A9. © 1993 San Francisco Chronicle. Reprinted by permission.

Solution
P (at least 1 will die)
1  P (all 3 will survive)
1  (50
51) 3
0.0577

EXAMPLE 2–7 A marketing research firm is interested in interviewing a consumer who fits certain
qualifications, for example, use of a certain product. The firm knows that 10% of
the public in a certain area use the product and would thus qualify to be inter-
viewed. The company selects a random sample of 10 people from the population as
a whole. What is the probability that at least 1 of these 10 people qualifies to be
interviewed?

Solution First, we note that if a sample is drawn at random, then the event that any one of the
items in the sample fits the qualifications is independent of the other items in the sam-
ple. This is an important property in statistics. Let Q i , where i
1, 2, . . . , 10, be the
event that person i qualifies. Then the probability that at least 1 of the 10 people will
qualify is the probability of the union of the 10 events Q i (i
1, . . . , 10). We are thus
looking for P(Q 1  Q 2      Q 10).
Now, since 10% of the people qualify, the probability that person i does not
qualify, or P ( Q i ), is equal to 0.90 for each i
1, . . . , 10. Therefore, the required
probability is equal to 1  (0.9)(0.9)    (0.9) (10 times), or 1  (0.9)10. This is equal
to 0.6513.

Be sure that you understand the difference between independent events and
mutually exclusive events. Although these two concepts are very different, they often
cause some confusion when introduced. When two events are mutually exclusive,
they are not independent. In fact, they are dependent events in the sense that if one
happens, the other one cannot happen. The probability of the intersection of two
mutually exclusive events is equal to zero. The probability of the intersection of
two independent events is not zero; it is equal to the product of the probabilities of
the separate events.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 71
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 69

PROBLEMS

2–36. According to USA Today, 65% of Americans are overweight or obese.8 If five
Americans are chosen at random, what is the probability that at least one of them is
overweight or obese?
2–37. The chancellor of a state university is applying for a new position. At a certain
point in his application process, he is being considered by seven universities. At three
of the seven he is a finalist, which means that (at each of the three universities) he is in
the final group of three applicants, one of which will be chosen for the position. At two
of the seven universities he is a semifinalist, that is, one of six candidates (in each of the
two universities). In two universities he is at an early stage of his application and
believes there is a pool of about 20 candidates for each of the two positions. Assuming
that there is no exchange of information, or influence, across universities as to their
hiring decisions, and that the chancellor is as likely to be chosen as any other appli-
cant, what is the chancellor’s probability of getting at least one job offer?
2–38. A package of documents needs to be sent to a given destination, and deliv-
ery within one day is important. To maximize the chances of on-time delivery, three
copies of the documents are sent via three different delivery services. Service A is
known to have a 90% on-time delivery record, service B has an 88% on-time delivery
record, and service C has a 91% on-time delivery record. What is the probability that
at least one copy of the documents will arrive at its destination on time?
2–39. The projected probability of increase in online holiday sales from 2004 to
2005 is 95% in the United States, 90% in Australia, and 85% in Japan. Assume these
probabilities are independent. What is the probability that holiday sales will increase
in all three countries from 2004 to 2005?
2–40. An electronic device is made up of two components A and B such that the
device would work satisfactorily as long as at least one of the components works. The
probability of failure of component A is 0.02 and that of B is 0.1 in some fixed period
of time. If the components work independently, find the probability that the device
will work satisfactorily during the period.
2–41. A recent survey conducted by Towers Perrin and published in the Financial
Times showed that among 460 organizations in 13 European countries, 93% have
bonus plans, 55% have cafeteria-style benefits, and 70% employ home-based work-
ers. If the types of benefits are independent, what is the probability that an organiza-
tion selected at random will have at least one of the three types of benefits?
2–42. Credit derivatives are a new kind of investment instrument: they protect
investors from risk.9 If such an investment offered by ABN Amro has a 90% chance
of making money, another by AXA has a 75% chance of success, and one by the ING
Group has a 60% chance of being profitable, and the three are independent of each
other, what is the chance that at least one investment will make money?
2–43. In problem 2–42, suppose that American investment institutions enter this
new market, and that their probabilities for successful instruments are:

Goldman Sachs 70%

Salomon Brothers 82%
Fidelity 80%
Smith Barney 90%

What is the probability that at least one of these four instruments is successful?
Assume independence.

8
Nancy Hellmich, “A Nation of Obesity,” USA Today, October 14, 2003, p. 7D.
9
John Ferry, “Gimme Shelter,” Worth, April 2007, pp. 88–90.
72 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

70 Chapter 2

2–44. In problem 2–31, are the events “hospitalization” and “the claim being from
the Midwest” independent of each other?
2–45. In problem 2–33, are “dividends paid” and “price increase” independent events?
2–46. In problem 2–34, are the events “top executive made more than $1 million”
and “shareholders lost money” independent of each other? If this is true for all firms,
how would you interpret your finding?
2–47. The accompanying table shows the incidence of malaria and two other similar
illnesses. If a person lives in an area affected by all three diseases, what is the proba-
bility that he or she will develop at least one of the three illnesses? (Assume that con-
tracting one disease is an event independent from contracting any other disease.)
Cases Number at Risk (Millions)
Malaria 110 million per year 2,100
Schistosomiasis 200 million 600
Sleeping sickness 25,000 per year 50

2–48. A device has three components and works as long as at least one of the com-
ponents is functional. The reliabilities of the components are 0.96, 0.91, and 0.80.
What is the probability that the device will work when needed?
2–49. In 2003, there were 5,732 deaths from car accidents in France.10 The popula-
tion of France is 59,625,919. If I am going to live in France for five years, what is my
probability of dying in a car crash?
2–50. The probabilities that three drivers will be able to drive home safely after
drinking are 0.5, 0.25, and 0.2, respectively. If they set out to drive home after drink-
ing at a party, what is the probability that at least one driver drives home safely?
2–51. When one is randomly sampling four items from a population, what is the
probability that all four elements will come from the top quartile of the population
distribution? What is the probability that at least one of the four elements will come
from the bottom quartile of the distribution?

2–6 Combinatorial Concepts

In this section we briefly discuss a few combinatorial concepts and give some formulas
useful in the analysis. The interested reader may find more on combinatorial rules and
their applications in the classic book by W. Feller or in other books on probability.11
If there are n events and event i can occur in Ni possible ways, then the num-
ber of ways in which the sequence of n events may occur is N1 N2    Nn.

Suppose that a bank has two branches, each branch has two departments, and each
department has three employees. Then there are (2)(2)(3) choices of employees, and
the probability that a particular one will be randomly selected is 1
(2)(2)(3)
1
12.
We may view the choice as done sequentially: First a branch is randomly chosen,
then a department within the branch, and then the employee within the department.
This is demonstrated in the tree diagram in Figure 2–8.

For any positive integer n, we define n factorial as

n(n  1)(n  2)    1
We denote n factorial by n!. The number n! is the number of ways in which
n objects can be ordered. By definition, 0!
1.

10
Elaine Sciolino, “Garçon! The Check, Please, and Wrap Up the Bordelais!,” The New York Times, January 26, 2004, p. A4.
11
William Feller, An Introduction to Probability Theory and Its Applications, vol. I, 3d ed. (New York: John Wiley & Sons, 1968).
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 73
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 71

FIGURE 2–8 Tree Diagram for Computing the Total Number of Employees by Multiplication

Employee:
Department: 1
2
Branch: 1
3
1
1 2 2
3
1
2
2 1
3
1
2 2
3
Total: 12

For example, 6! is the number of possible arrangements of six objects. We have 6!

(6)(5)(4)(3)(2)(1)
720. Suppose that six applications arrive at a center on the same
day, all written at different times. What is the probability that they will be read in the
order in which they were written? Since there are 720 ways to order six applications,
the probability of a particular order (the order in which the applications were written)
is 1
720.

Permutations are the possible ordered selections of r objects out of a total

of n objects. The number of permutations of n objects taken r at a time is
denoted nPr.
n!
nPr
(2–12)
(n  r)!

Suppose that 4 people are to be randomly chosen out of 10 people who agreed to be
interviewed in a market survey. The four people are to be assigned to four inter-
viewers. How many possibilities are there? The first interviewer has 10 choices, the
second 9 choices, the third 8, and the fourth 7. Thus, there are (10)(9)(8)(7)
5,040
selections. You can see that this is equal to n(n  1)(n  2)    (n  r  1), which is
equal to n!
(n  r)!. If choices are made randomly, the probability of any predeter-
mined assignment of 4 people out of a group of 10 is 1
5,040.

Combinations are the possible selections of r items from a group of n items

regardless of the order of selection. The number of combinations is denoted
by (nr) and is read n choose r. An alternative notation is nCr. We define the
number of combinations of r out of n elements as
n n!
¢r≤
(2–13)
r!(n  r)!

This is the most important of the combinatorial rules given in this chapter and
is the only one we will use extensively. This rule is basic to the formula of the
binomial distribution presented in the next chapter and will find use also in
other chapters.
74 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

72 Chapter 2

Suppose that 3 out of the 10 members of the board of directors of a large corpo-
ration are to be randomly selected to serve on a particular task committee. How
many possible selections are there? Using equation 2–13, we find that the number of
combinations is (103 )
10!
(3!7!)
120. If the committee is chosen in a truly random
fashion, what is the probability that the three committee members chosen will be the
three senior board members? This is 1 combination out of a total of 120, so the
answer is 1
120
0.00833.

EXAMPLE 2–8 A certain university held a meeting of administrators and faculty members to discuss
some important issues of concern to both groups. Out of eight members, two were
faculty, and both were missing from the meeting. If two members are absent, what is
the probability that they should be the two faculty members?

Solution By definition, there are (82 ) ways of selecting two people out of a total of eight people,
disregarding the order of selection. Only one of these ways corresponds to the pair’s
being the two faculty members. Hence, the probability is 1
(28)
1
[8!
(2!6!)]
1
28

0.0357. This assumes randomness.

PROBLEMS

2–52. A company has four departments: manufacturing, distribution, marketing,

and management. The number of people in each department is 55, 30, 21, and 13,
respectively. Each department is expected to send one representative to a meeting
with the company president. How many possible sets of representatives are there?
2–53. Nine sealed bids for oil drilling leases arrive at a regulatory agency in the
morning mail. In how many different orders can the nine bids be opened?
2–54. Fifteen locations in a given area are believed likely to have oil. An oil
company can only afford to drill at eight sites, sequentially chosen. How many
possibilities are there, in order of selection?
2–55. A committee is evaluating six equally qualified candidates for a job. Only
three of the six will be invited for an interview; among the chosen three, the order of
invitation is of importance because the first candidate will have the best chance of
being accepted, the second will be made an offer only if the committee rejects the
first, and the third will be made an offer only if the committee should reject both the
first and the second. How many possible ordered choices of three out of six candidates
are there?
2–56. In the analysis of variance (discussed in Chapter 9) we compare several pop-
ulation means to see which is largest. After the primary analysis, pairwise compar-
isons are made. If we want to compare seven populations, each with all the others,
how many pairs are there? (We are looking for the number of choices of seven items
taken two at a time, regardless of order.)
2–57. In a shipment of 14 computer parts, 3 are faulty and the remaining 11 are in
working order. Three elements are randomly chosen out of the shipment. What is the
probability that all three faulty elements will be the ones chosen?
2–58. Megabucks is a lottery game played in Massachusetts with the following
rules. A random drawing of 6 numbers out of all 36 numbers from 1 to 36 is made
every Wednesday and every Saturday. The game costs $1 to play, and to win a
person must have the correct six numbers drawn, regardless of their order. (The
numbers are sequentially drawn from a bin and are arranged from smallest to
largest. When a player buys a ticket prior to the drawing, the player must also
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 75
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 73

arrange his or her chosen numbers in ascending order.) The jackpot depends on
the number of players and is usually worth several million dollars. What is the
probability of winning the jackpot?
2–59. In Megabucks, a player who correctly chooses five out of the six winning
numbers gets $400. What is the probability of winning $400?

2–7 The Law of Total Probability

and Bayes’ Theorem
In this section we present two useful results of probability theory. The first one, the
law of total probability, allows us at times to evaluate probabilities of events that are
difficult to obtain alone, but become easy to calculate once we condition on the occur-
rence of a related event. First we assume that the related event occurs, and then we
assume it does not occur. The resulting conditional probabilities help us compute the
total probability of occurrence of the event of interest.
The second rule, the famous Bayes’ theorem, is easily derived from the law of
total probability and the definition of conditional probability. The rule, discovered
in 1761 by the English clergyman Thomas Bayes, has had a profound impact on the
development of statistics and is responsible for the emergence of a new philosophy
of science. Bayes himself is said to have been unsure of his extraordinary result,
which was presented to the Royal Society by a friend in 1763—after Bayes’ death.

The Law of Total Probability

Consider two events A and B. Whatever may be the relation between the two
events, we can always say that the probability of A is equal to the probability of the
intersection of A and B, plus the probability of the intersection of A and the comple-
ment of B (event B).

The law of total probability:

FIGURE 2–9
P(A)
P(A ¨ B)  P(A ¨ B) (2–14) Partition of Set A into Its
Intersections with the
Two Sets B and B, and the
Implied Law of Total
Probability
The sets B and B form a partition of the sample space. A partition of a space is the
division of the space into a set of events that are mutually exclusive (disjoint sets) and
P(A) = P(A
B) + P(A
B)
cover the whole space. Whatever event B may be, either B or B must occur, but not
both. Figure 2–9 demonstrates this situation and the law of total probability.
The law of total probability may be extended to more complex situations, where the B
sample space X is partitioned into more than two events. Say we partition the space into
a collection of n sets B1, B2, . . . , Bn. The law of total probability in this situation is
A
B

n
P(A)
a P (A ¨ Bi) (2–15)
i
1

A
B

Figure 2–10 shows the partition of a sample space into the four events B1, B2, B3, and
B4 and shows their intersections with set A.
We demonstrate the rule with a more specific example. Define A as the event that B
a picture card is drawn out of a deck of 52 cards (the picture cards are the aces, kings,
queens, and jacks). Letting H, C, D, and S denote the events that the card drawn is a X A
heart, club, diamond, or spade, respectively, we find that the probability of a picture
76 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

74 Chapter 2

FIGURE 2–10 The Partition of Set A into Its Intersection with Four Partition Sets

B1 B2 B3 B4

A
B1 A
B2 A
B3 A
B4

FIGURE 2–11 The Total Probability of Drawing a Picture Card as the Sum of the Probabilities
of Drawing a Card in the Intersections of Picture and Suit

H D C S

A A A A
K K K K
A
H Q Q Q Q Event A
J J J J A
S
A
D 10 10 10 10
9 9 9 9 A
C
8 8 8 8
7 7 7 7
6 6 6 6 X
5 5 5 5
4 4 4 4
3 3 3 3
2 2 2 2

card is P(A)
P(A
H)  P(A
C)  P (A
D)  P (A
S)
4
52  4
52 

4
52  4
52
16
52, which is what we know the probability of a picture card to be
just by counting 16 picture cards out of a total of 52 cards in the deck. This demon-
strates equation 2–15. The situation is shown in Figure 2–11. As can be seen from the
figure, the event A is the set addition of the intersections of A with each of the four
sets H, D, C, and S. Note that in these examples we denote the sample space X.
The law of total probability can be extended by using the definition of condition-
al probability. Recall that P (A
B)
P (A | B)P (B) (equation 2–8) and, similarly,
P (A ¨ B)
P (A ƒ B)P (B). Substituting these relationships into equation 2–14 gives
us another form of the law of total probability. This law and its extension to a parti-
tion consisting of more than two sets are given in equations 2–16 and 2–17. In equa-
tion 2–17, we have a set of conditioning events Bi that span the entire sample space,
instead of just two events spanning it, B and B.

The law of total probability using conditional probabilities:

Two-set case:

P (A)
P(A ƒ B)P (B)  P (A ƒ B)P (B) (2–16)

Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 77
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 75

More than two sets in the partition:

n
P(A)
a P(A ƒ Bi)P (Bi) (2–17)
i
1

where there are n sets in the partition: Bi, i
1, . . . , n.

An analyst believes the stock market has a 0.75 probability of going up in the next EXAMPLE 2–9
year if the economy should do well, and a 0.30 probability of going up if the economy
should not do well during the year. The analyst further believes there is a 0.80 prob-
ability that the economy will do well in the coming year. What is the probability that
the stock market will go up next year (using the analyst’s assessments)?

We define U as the event that the market will go up and W as the event the economy Solution
will do well. Using equation 2–16, we find P(U)
P(U | W)P(W)  P(U | W)P(W)

(0.75)(0.80)  (0.30)(0.20)
0.66.

Bayes’ Theorem
We now develop the well-known Bayes’ theorem. The theorem allows us to reverse
the conditionality of events: we can obtain the probability of B given A from the
probability of A given B (and other information).
By the definition of conditional probability, equation 2–7,

P (A ¨ B)
P(B ƒ A)
(2–18)
P (A)

By another form of the same definition, equation 2–8,

P(A
B)
P(A | B)P(B) (2–19)

Substituting equation 2–19 into equation 2–18 gives

P (A ƒ B)P (B)
P(B ƒ A)
(2–20)
P (A)

From the law of total probability using conditional probabilities, equation 2–16, we have

P(A)
P(A ƒ B)P(B)  P (A ƒ B)P (B)

Substituting this expression for P(A) in the denominator of equation 2–20 gives us
Bayes’ theorem.
78 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

76 Chapter 2

Bayes’ Theorem
P(A ƒ B)P (B)
P (B ƒ A)
(2–21)
P (A ƒ B)P (B)  P (A ƒ B)P (B)

As we see from the theorem, the probability of B given A is obtained from the prob-
abilities of B and B and from the conditional probabilities of A given B and A given B.
The probabilities P (B) and P (B) are called prior probabilities of the events B
and B; the probability P(B | A) is called the posterior probability of B. Bayes’
theorem may be written in terms of B and A, thus giving the posterior probability of
B, P( B | A). Bayes’ theorem may be viewed as a means of transforming our prior
probability of an event B into a posterior probability of the event B—posterior to the
known occurrence of event A.
The use of prior probabilities in conjunction with other information—often
obtained from experimentation—has been questioned. The controversy arises in
more involved statistical situations where Bayes’ theorem is used in mixing the objec-
tive information obtained from sampling with prior information that could be subjec-
tive. We will explore this topic in greater detail in Chapter 15. We now give some
examples of the use of the .

EXAMPLE 2–10 Consider a test for an illness. The test has a known reliability:

1. When administered to an ill person, the test will indicate so with probability 0.92.
2. When administered to a person who is not ill, the test will erroneously give a
positive result with probability 0.04.
Suppose the illness is rare and is known to affect only 0.1% of the entire population.
If a person is randomly selected from the entire population and is given the test and
the result is positive, what is the posterior probability (posterior to the test result) that
the person is ill?

Solution Let Z denote the event that the test result is positive and I the event that the person
tested is ill. The preceding information gives us the following probabilities of events:

P(I)
0.001 P (I)
0.999 P (Z ƒ I)
0.92 P (Z ƒ I)
0.04

We are looking for the probability that the person is ill given a positive test result; that
is, we need P(I | Z). Since we have the probability with the reversed conditionality,
P(Z | I), we know that Bayes’ theorem is the rule to be used here. Applying the rule,
equation 2–21, to the events Z, I, and I, we get

P (Z ƒ I)P (I) (0.92)(0.001)

P(I ƒ Z)


P (Z ƒ I)P (I)  P (Z ƒ I)P (I) (0.92)(0.001)  (0.04)(0.999)

0.0225

This result may surprise you. A test with a relatively high reliability (92% correct
diagnosis when a person is ill and 96% correct identification of people who are not ill)
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 79
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 77

is administered to a person, the result is positive, and yet the probability that the per-
son is actually ill is only 0.0225!
The reason for the low probability is that we have used two sources of information
here: the reliability of the test and the very small probability (0.001) that a randomly
selected person is ill. The two pieces of information were mixed by Bayes’ theorem,
and the posterior probability reflects the mixing of the high reliability of the test with
the fact that the illness is rare. The result is perfectly correct as long as the informa-
tion we have used is accurate. Indeed, subject to the accuracy of our information,
if the test were administered to a large number of people selected randomly from the
entire population, it would be found that about 2.25% of the people in the sample
who test positive are indeed ill.
Problems with Bayes’ theorem arise when we are not careful with the use of prior
information. In this example, suppose the test is administered to people in a hospital.
Since people in a hospital are more likely to be ill than people in the population as a
whole, the overall population probability that a person is ill, 0.001, no longer applies.
If we applied this low probability in the hospital, our results would not be correct.
This caution extends to all situations where prior probabilities are used: We must
always examine the appropriateness of the prior probabilities.

Bayes’ theorem may be extended to a partition of more than two sets. This is
done using equation 2–17, the law of total probability involving a partition of sets B1,
B2, . . . , Bn. The resulting extended form of Bayes’ theorem is given in equation 2–22.
The theorem gives the probability of one of the sets in partition B1 given the occur-
rence of event A. A similar expression holds for any of the events Bi.

Extended Bayes’ Theorem

P (A ƒ B1)P(B1)
P(B1 ƒ A)
n (2–22)
a P (A ƒ Bi)P(Bi)
i
1

We demonstrate the use of equation 2–22 with the following example. In the solution,
we use a table format to facilitate computations. We also demonstrate the computations
using a tree diagram.

An economist believes that during periods of high economic growth, the U.S. dollar EXAMPLE 2–11
appreciates with probability 0.70; in periods of moderate economic growth, the dollar
appreciates with probability 0.40; and during periods of low economic growth, the
dollar appreciates with probability 0.20. During any period of time, the probability of
high economic growth is 0.30, the probability of moderate growth is 0.50, and the
probability of low economic growth is 0.20. Suppose the dollar has been appreciating
during the present period. What is the probability we are experiencing a period of
high economic growth?

Figure 2–12 shows solution by template. Below is the manual solution. Solution
Our partition consists of three events: high economic growth (event H), moderate
economic growth (event M), and low economic growth (event L). The prior probabil-
ities of the three states are P(H)
0.30, P(M)
0.50, and P(L)
0.20. Let A denote
80 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

78 Chapter 2

FIGURE 2–12 Bayesian Revision of Probabilities

[Bayes Revision.xls; Sheet: Empirical]

A B C D E F G H I J K L M
1 Bayesian Revision based on Empirical Conditional Probabilities Example 2-11
2
3 High Moderate Low
4 s1 s2 s3 s4 s5 s6 s7 s8 Total
5 Prior Probability 0.3 0.5 0.2 1
6
7 Conditional Probabilities
8 s1 s2 s3 s4 s5 s6 s7 s8
9 $ Appreciates P(l1 | .) 0.7 0.4 0.2
10 $ Depreciates P(l2 | .) 0.3 0.6 0.8
11 P(l3 | .)
12 P(l4 | .)
13 P(l5 | .)
14 Total 1 1 1 0 0 0 0 0
15
16 Joint Probabilities
17 s1 s2 s3 s4 s5 s6 s7 s8 Marginal
18 l1 0.2100 0.2000 0.0400 0.4500
19 l2 0.0900 0.3000 0.1600 0.5500
20 l3
21 l4
22 l5
23
24 Posterior Probabilities
25 s1 s2 s3 s4 s5 s6 s7 s8
26 P(. | l1) 0.4667 0.4444 0.0889
27 P(. | l2) 0.1636 0.5455 0.2909
28 P(. | l3)
29 P(. | l4)
30 P(. | l5)

the event that the dollar appreciates. We have the following conditional probabilities:
P(A | H)
0.70, P(A | M)
0.40, and P(A | L)
0.20. Applying equation 2–22 while
using three sets (n
3), we get

P (A ƒ H)P (H)
P(H ƒ A)

P (A ƒ H)P(H)  P (A ƒ M)P (M)  P (A ƒ L)P (L)
(0.70)(0.30)


0.467
(0.70)(0.30)  (0.40)(0.50)  (0.20)(0.20)

We can obtain this answer, along with the posterior probabilities of the other two
states, M and L, by using a table. In the first column of the table we write the prior
probabilities of the three states H, M, and L. In the second column we write the three
conditional probabilities P (A | H), P (A | M), and P (A | L). In the third column we write
the joint probabilities P (A
H), P(A
M), and P (A
L). The joint probabilities
are obtained by multiplying across in each of the three rows (these operations make
use of equation 2–8). The sum of the entries in the third column is the total proba-
bility of event A (by equation 2–15). Finally, the posterior probabilities P (H | A),
P(M | A), and P(L | A) are obtained by dividing the appropriate joint probability by
the total probability of A at the bottom of the third column. For example, P(H | A) is
obtained by dividing P (H
A) by the probability P (A). The operations and the
results are given in Table 2–1 and demonstrated in Figure 2–13.
Note that both the prior probabilities and the posterior probabilities of the
three states add to 1.00, as required for probabilities of all the possibilities in a
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 81
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 79

TABLE 2–1 Bayesian Revision of Probabilities for Example 2–11

Prior Conditional Joint Posterior

Event Probability Probability Probability Probability
0.21
H P(H)
0.30 P (A | H)
0.70 P (A
H)
0.21 P (H | A)

0.467
0.45
0.20
M P(M)
0.50 P (A | M)
0.40 P(A
M)
0.20 P (M | A)

0.444
0.45
0.04
L P (L)
0.20 P(A | L)
0.20 P (A
L)
0.04 P (L | A)

0.089
0.45
Sum
1.00 P (A)
0.45 Sum
1.000

FIGURE 2–13 Tree Diagram for Example 2–11

Conditional Joint probabilities:

probabilities: (by multiplication)
Prior
P (A I H) = 0.70
Probabilities: P(H
A) = (0.30) (0.70) = 0.21

P(H) = 0.30 P(A I H) = 0.30

P (H
A) = (0.30) (0.30) = 0.09
P(A I M) = 0.40
P (M
A) = (0.50) (0.40) = 0.20
P(M) = 0.50
P(A I M) = 0.60
P(M
A) = (0.50) (0.60) = 0.30
P(A I L) = 0.20
P (L
A) = (0.20) (0.20) = 0.04
P(L) = 0.20
Sum = 1.00 P(A I L) = 0.80
P (L
A) = (0.20) (0.80) = 0.16

given situation. We conclude that, given that the dollar has been appreciating, the
probability that our period is one of high economic growth is 0.467, the probability
that it is one of moderate growth is 0.444, and the probability that our period is one
of low economic growth is 0.089. The advantage of using a table is that we can
obtain all posterior probabilities at once. If we use the formula directly, we need to
apply it once for the posterior probability of each state.

2–8 The Joint Probability Table

A joint probability table is similar to a contingency table, except that it has proba-
bilities in place of frequencies. For example, the case in Example 2–11 can be sum-
marized with the joint probabilities shown in Table 2–2. The body of the table can
be visualized as the sample space partitioned into row-events and column-events.

TABLE 2–2 Joint Probability Table

High Medium Low Total

$ Appreciates 0.21 0.2 0.04 0.45
$ Depreciates 0.09 0.3 0.16 0.55
Total 0.3 0.5 0.2 1
82 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

80 Chapter 2

Every cell is a joint event of that row and column. Thus the joint probability of High
and $ Appreciates is 0.21.
The row totals and column totals are known as marginal probabilities. For exam-
ple, the marginal probability of the event “High” is 0.3. In Example 2–11, this was the
prior probability of High. The marginal probability of the event “$ Appreciates” is 0.45.
The computations shown in Table 2–1 yield the top row of values in the joint probabili-
ty table. Knowing the column totals (which are the prior or marginal probabilities of
High, Medium, and Low), we can quickly infer the second row of values.
The joint probability table is also a clear means to compute conditional proba-
bilities. For example, the conditional probability that $ Appreciates when economic
growth is High can be computed using the Bayes’ formula:

P ($ Appreciates and High)

P ($ Appreciates |High) =
P (High)

Note that the numerator in the formula is a joint probability and the denominator is
a marginal probability.

0.21
P ($ Appreciates | High) = = 0.467
0.45

which is the posterior probability sought in Example 2–11.

If you use the template Bayesian Revision.xls to solve a problem, you will note
that the template produces the joint probability table in the range C18:J22. It also
computes all marginal and all conditional probabilities.

2–9 Using the Computer

Excel Templates and Formulas
Figure 2–14 shows the template which is used for calculating joint, marginal, and con-
ditional probabilities starting from a contingency table. If the starting point is a joint
probability table, rather than a contingency table, this template can still be used.
Enter the joint probability table in place of the contingency table.
The user needs to know how to read off the conditional probabilities from this
template. The conditional probability of 0.6667 in cell C23 is P (Telecom| AT&T),
which is the row label in cell B23 and the column label in cell C22 put together.
Similarly, the conditional probability of 0.8000 in cell K14 is P (AT&T | Telecom).
Figure 2–12 shows the template that can be used to solve conditional probability
problems using Bayes’ revision of probabilities. It was used to solve Example 2–11.
In addition to the template mentioned above, you can also directly use Excel func-
tions for some of the calculations in this chapter. For example, functions COMBIN
(number of items, number to choose) and PERMUT (number of items, number to
choose) are used to provide us with the number of combinations and permutations
of the number of items chosen some at a time. The function FACT (number) also
returns the factorial of a number. The numeric arguments of all these functions should
be nonnegative, and they will be truncated if they are not integers. Note that the
entries in the range C10:E10 of probabilities of the dollar depreciating have been
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 83
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 81

FIGURE 2–14 Template for Calculating Probabilities from a Contingency Table

[Contingency Table.xls]

A B C D E F G H I J K L M
1 Contingency Table and Conditional Probabilities Title
2
3 Contingency Table
4 AT&T IBM Total
5 Telecom 40 10 50
6 Comp 20 30 50
7
8
9
10 Total 60 40 100
11
12 Joint Probabilities Row-Conditional Probabilities
13 AT&T IBM Marginal P(AT&T P(IBM
14 Telecom 0.4000 0.1000 0.5000 |Telecom) 0.8000 0.2000
15 Comp 0.2000 0.3000 0.5000 |Comp) 0.4000 0.6000
16
17
18
19 Marginal 0.6000 0.4000 1.0000
20
21 Column-Conditional Probabilities
22 |AT&T) |IBM)
23 P(Telecom 0.6667 0.2500
24 P(comp 0.3333 0.7500
25
26
27
28

entered for completeness. The questions in the example can be answered even with-
out those entries.

Using MINITAB
We can use MINITAB to find a large number of arithmetic operations such as factorial,
combination, and permutation. The command Let C1 = FACTORIAL(n) calcu-
lates n factorial (n !), the product of all the consecutive integers from 1 to n inclusive,
and puts the result in the first cell of column C1. The value of n (number of items)
must be greater than or equal to 0. You can enter a column or constant and missing
values are not allowed. You can also use the menu by choosing Calc
Calculator. In
the list of functions choose FACTORIAL and then specify the number of items. You
need also define the name of the variable that will store the result, for example, C1,
then press OK.
The command Let C1 = COMBINATIONS(n,k) calculates the number of com-
binations of n items chosen k at a time. You can specify the number of items (n) and
the number to choose (k) as columns or constants. The number of items must be
greater than or equal to 1, and the number to choose must be greater than or equal
to 0. Missing values are not allowed. The same as before, you can use menu Calc

Calculator and choose COMBINATIONS in the list of functions. Then specify the number
of items, the number to choose, and the name of the variable that will store the results.
Then press OK.
The next command is Let C1 = PERMUTATIONS(n,k), which calculates the
number of permutations of n things taken k at a time. Specify the number of items
(n) and the number to choose (k). The number of items must be greater than or
equal to 1, and the number to choose must be greater than or equal to 0. Missing
values are not allowed.
Figure 2–15 shows how we can use session commands and the menu to obtain
permutations and combinations.
84 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

82 Chapter 2

FIGURE 2–15 Using MINITAB for Permutation and Combination Problems

PROBLEMS

2–60. In a takeover bid for a certain company, management of the raiding firm
believes that the takeover has a 0.65 probability of success if a member of the
board of the raided firm resigns, and a 0.30 chance of success if she does not
resign. Management of the raiding firm further believes that the chances for a
resignation of the member in question are 0.70. What is the probability of a suc-
cessful takeover?
2–61. A drug manufacturer believes there is a 0.95 chance that the Food and Drug
Administration (FDA) will approve a new drug the company plans to distribute if
the results of current testing show that the drug causes no side effects. The manu-
facturer further believes there is a 0.50 probability that the FDA will approve the
drug if the test shows that the drug does cause side effects. A physician working for
the drug manufacturer believes there is a 0.20 probability that tests will show that
the drug causes side effects. What is the probability that the drug will be approved
by the FDA?
2–62. An import–export firm has a 0.45 chance of concluding a deal to export agri-
cultural equipment to a developing nation if a major competitor does not bid for the
contract, and a 0.25 probability of concluding the deal if the competitor does bid for
it. It is estimated that the competitor will submit a bid for the contract with probabil-
ity 0.40. What is the probability of getting the deal?
2–63. A realtor is trying to sell a large piece of property. She believes there is a 0.90
probability that the property will be sold in the next 6 months if the local economy
continues to improve throughout the period, and a 0.50 probability the property will
be sold if the local economy does not continue its improvement during the period. A
state economist consulted by the realtor believes there is a 0.70 chance the economy
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 85
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 83

will continue its improvement during the next 6 months. What is the probability that
the piece of property will be sold during the period?
2–64. Holland America Cruise Lines has three luxury cruise ships that sail to Alaska
during the summer months. Since the business is very competitive, the ships must run
full during the summer if the company is to turn a profit on this line. A tourism expert
hired by Holland America believes there is a 0.92 chance the ships will sail full during
the coming summer if the dollar does not appreciate against European currencies, and
a 0.75 chance they will sail full if the dollar does appreciate in Europe (appreciation of
the dollar in Europe draws U.S. tourists there, away from U.S. destinations). Econo-
mists believe the dollar has a 0.23 chance of appreciating against European currencies
soon. What is the probability the ships will sail full?
2–65. Saflok is an electronic door lock system made in Troy, Michigan, and used
in modern hotels and other establishments. To open a door, you must insert the elec-
tronic card into the lock slip. Then a green light indicates that you can turn the
handle and enter; a yellow light indicates that the door is locked from inside, and
you cannot enter. Suppose that 90% of the time when the card is inserted, the door
should open because it is not locked from inside. When the door should open, a
green light will appear with probability 0.98. When the door should not open,
a green light may still appear (an electronic error) 5% of the time. Suppose that you
just inserted the card and the light is green. What is the probability that the door will
actually open?
2–66. A chemical plant has an emergency alarm system. When an emergency sit-
uation exists, the alarm sounds with probability 0.95. When an emergency situation
does not exist, the alarm system sounds with probability 0.02. A real emergency sit-
uation is a rare event, with probability 0.004. Given that the alarm has just sounded,
what is the probability that a real emergency situation exists?
2–67. When the economic situation is “high,” a certain economic indicator rises
with probability 0.6. When the economic situation is “medium,” the economic indi-
cator rises with probability 0.3. When the economic situation is “low,” the indicator
rises with probability 0.1. The economy is high 15% of the time, it is medium 70% of
the time, and it is low 15% of the time. Given that the indicator has just gone up, what
is the probability that the economic situation is high?
2–68. An oil explorer orders seismic tests to determine whether oil is likely to be
found in a certain drilling area. The seismic tests have a known reliability: When oil
does exist in the testing area, the test will indicate so 85% of the time; when oil does
not exist in the test area, 10% of the time the test will erroneously indicate that it does
exist. The explorer believes that the probability of existence of an oil deposit in the
test area is 0.4. If a test is conducted and indicates the presence of oil, what is the
probability that an oil deposit really exists?
2–69. Before marketing new products nationally, companies often test them on
samples of potential customers. Such tests have a known reliability. For a particu-
lar product type, a test will indicate success of the product 75% of the time if
the product is indeed successful and 15% of the time when the product is not suc-
cessful. From past experience with similar products, a company knows that a
new product has a 0.60 chance of success on the national market. If the test indi-
cates that the product will be successful, what is the probability that it really will
be successful?
2–70. A market research field worker needs to interview married couples about
use of a certain product. The researcher arrives at a residential building with three
apartments. From the names on the mailboxes downstairs, the interviewer infers
that a married couple lives in one apartment, two men live in another, and two
women live in the third apartment. The researcher goes upstairs and finds that there
are no names or numbers on the three doors, so that it is impossible to tell in which
86 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

84 Chapter 2

of the three apartments the married couple lives. The researcher chooses a door at
random and knocks. A woman answers the door. Having seen a woman at the
door, what now is the probability of having reached the married couple? Make
the (possibly unrealistic) assumptions that if the two men’s apartment was reached,
a woman cannot answer the door; if the two women’s apartment was reached,
then only a woman can answer; and that if the married couple was reached, then
the probability of a woman at the door is 1
2. Also assume a 1
3 prior probability
of reaching the married couple. Are you surprised by the numerical answer you
obtained?

2–10 Summary and Review of Terms

In this chapter, we discussed the basic ideas of probability. We defined probability as
a relative measure of our belief in the occurrence of an event. We defined a sample
space as the set of all possible outcomes in a given situation and saw that an event is a
set within the sample space. We set some rules for handling probabilities: the rule of
unions, the definition of conditional probability, the law of total probability, and
Bayes’ theorem. We also defined mutually exclusive events and independence of
events. We saw how certain computations are possible in the case of independent
events, and we saw how we may test whether events are independent.
In the next chapter, we will extend the ideas of probability and discuss random
variables and probability distributions. These will bring us closer to statistical infer-
ence, the main subject of this book.

PROBLEMS

2–71. AT&T was running commercials in 1990 aimed at luring back customers who
had switched to one of the other long-distance phone service providers. One such
commercial shows a businessman trying to reach Phoenix and mistakenly getting
Fiji, where a half-naked native on a beach responds incomprehensibly in Polynesian.
When asked about this advertisement, AT&T admitted that the portrayed incident
did not actually take place but added that this was an enactment of something that
“could happen.”12 Suppose that one in 200 long-distance telephone calls is misdirect-
ed. What is the probability that at least one in five attempted telephone calls reaches
the wrong number? (Assume independence of attempts.)
2–72. Refer to the information in the previous problem. Given that your long-
distance telephone call is misdirected, there is a 2% chance that you will reach a for-
eign country (such as Fiji). Suppose that I am now going to dial a single long-distance
number. What is the probability that I will erroneously reach a foreign country?
2–73. The probability that a builder of airport terminals will win a contract for con-
struction of terminals in country A is 0.40, and the probability that it will win a con-
tract in country B is 0.30. The company has a 0.10 chance of winning the contracts in
both countries. What is the probability that the company will win at least one of these
two prospective contracts?
2–74. According to BusinessWeek, 50% of top managers leave their jobs within
5 years.13 If 25 top managers are followed over 5 years after they assume their posi-
tions, what is the probability that none will have left their jobs? All of them will have

12
While this may seem virtually impossible due to the different dialing procedure for foreign countries, AT&T argues
that erroneously dialing the prefix 679 instead of 617, for example, would get you Fiji instead of Massachusetts.
13
Roger O. Crockett, “At the Head of the Headhunting Pack,” BusinessWeek, April 9, 2007, p. 80.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 87
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 85

left their jobs? At least one will have left the position? What implicit assumption are
you making and how do you justify it?
2–75. The probability that a consumer entering a retail outlet for microcomputers
and software packages will buy a computer of a certain type is 0.15. The probability
that the consumer will buy a particular software package is 0.10. There is a 0.05 prob-
ability that the consumer will buy both the computer and the software package. What
is the probability that the consumer will buy the computer or the software package
or both?
2–76. The probability that a graduating senior will pass the certified public accountant
(CPA) examination is 0.60. The probability that the graduating senior will both pass
the CPA examination and get a job offer is 0.40. Suppose that the student just found
out that she passed the CPA examination. What is the probability that she will be
offered a job?
2–77. Two stocks A and B are known to be related in that both are in the same indus-
try. The probability that stock A will go up in price tomorrow is 0.20, and the proba-
bility that both stocks A and B will go up tomorrow is 0.12. Suppose that tomorrow
you find that stock A did go up in price. What is the probability that stock B went up
as well?
2–78. The probability that production will increase if interest rates decline more
than 0.5 percentage point for a given period is 0.72. The probability that interest rates
will decline by more than 0.5 percentage point in the period in question is 0.25. What
is the probability that, for the period in question, both the interest rate will decline
and production will increase?
2–79. A large foreign automaker is interested in identifying its target market in
the United States. The automaker conducts a survey of potential buyers of its high-
performance sports car and finds that 35% of the potential buyers consider engi-
neering quality among the car’s most desirable features and that 50% of the people
surveyed consider sporty design to be among the car’s most desirable features.
Out of the people surveyed, 25% consider both engineering quality and sporty
design to be among the car’s most desirable features. Based on this information, do
you believe that potential buyers’ perceptions of the two features are independent?
Explain.
2–80. Consider the situation in problem 2–79. Three consumers are chosen ran-
domly from among a group of potential buyers of the high-performance automobile.
What is the probability that all three of them consider engineering quality to be
among the most important features of the car? What is the probability that at least
one of them considers this quality to be among the most important ones? How do
you justify your computations?
2–81. A financial service company advertises its services in magazines, runs billboard
ads on major highways, and advertises its services on the radio. The company esti-
mates that there is a 0.10 probability that a given individual will see the billboard
ad during the week, a 0.15 chance that he or she will see the ad in a magazine, and a
0.20 chance that she or he will hear the advertisement on the radio during the week.
What is the probability that a randomly chosen member of the population in the area
will be exposed to at least one method of advertising during a given week? (Assume
independence.)
2–82. An accounting firm carries an advertisement in The Wall Street Journal.
The firm estimates that 60% of the people in the potential market read The Wall
Street Journal; research further shows that 85% of the people who read the Journal
remember seeing the advertisement when questioned about it afterward. What
percentage of the people in the firm’s potential market see and remember the
advertisement?
88 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

86 Chapter 2

2–83. A quality control engineer knows that 10% of the microprocessor chips pro-
duced by a machine are defective. Out of a large shipment, five chips are chosen at
random. What is the probability that none of them is defective? What is the proba-
bility that at least one is defective? Explain.
2–84. A fashion designer has been working with the colors green, black, and red in
preparing for the coming season’s fashions. The designer estimates that there is a 0.3
chance that the color green will be “in” during the coming season, a 0.2 chance that
black will be among the season’s colors, and a 0.15 chance that red will be popular.
Assuming that colors are chosen independently of each other for inclusion in new
fashions, what is the probability that the designer will be successful with at least one
of her colors?
2–85. A company president always invites one of her three vice presidents to
attend business meetings and claims that her choice of the accompanying vice pres-
ident is random. One of the three has not been invited even once in five meetings.
What is the probability of such an occurrence if the choice is indeed random? What
conclusion would you reach based on your answer?
2–86. A multinational corporation is considering starting a subsidiary in an Asian
country. Management realizes that the success of the new subsidiary depends, in part,
on the ensuing political climate in the target country. Management estimates that the
probability of success (in terms of resulting revenues of the subsidiary during its first
year of operation) is 0.55 if the prevailing political situation is favorable, 0.30 if the
political situation is neutral, and 0.10 if the political situation during the year is unfa-
vorable. Management further believes that the probabilities of favorable, neutral, and
unfavorable political situations are 0.6, 0.2, and 0.2, respectively. What is the success
probability of the new subsidiary?
2–87. The probability that a shipping company will obtain authorization to include
a certain port of call in its shipping route is dependent on whether certain legislation
is passed. The company believes there is a 0.5 chance that both the relevant legisla-
tion will pass and it will get the required authorization to visit the port. The company
further estimates that the probability that the legislation will pass is 0.75. If the com-
pany should find that the relevant legislation just passed, what is the probability that
authorization to visit the port will be granted?
2–88. The probability that a bank customer will default on a loan is 0.04 if the econ-
omy is high and 0.13 if the economy is not high. Suppose the probability that the
economy will be high is 0.65. What is the probability that the person will default on
the loan?
2–89. Researchers at Kurume University in Japan surveyed 225 workers aged
41 to 60 years and found that 30% of them were skilled workers and 70% were
unskilled. At the time of survey, 15% of skilled workers and 30% of unskilled work-
ers were on an assembly line. A worker is selected at random from the age group
41 to 60.
a. What is the probability that the worker is on an assembly line?
b. Given that the worker is on an assembly line, what is the probability that
the worker is unskilled?
2–90. SwissAir maintains a mailing list of people who have taken trips to Europe in
the last three years. The airline knows that 8% of the people on the mailing list will
make arrangements to fly SwissAir during the period following their being mailed
a brochure. In an experimental mailing, 20 people are mailed a brochure. What is
the probability that at least one of them will book a flight with SwissAir during the
coming season?
2–91. A company’s internal accounting standards are set to ensure that no more than
5% of the accounts are in error. From time to time, the company collects a random
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 89
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 87

sample of accounts and checks to see how many are in error. If the error rate is
indeed 5% and 10 accounts are chosen at random, what is the probability that none
will be in error?
2–92. At a certain university, 30% of the students who take basic statistics are first-
year students, 35% are sophomores, 20% are juniors, and 15% are seniors. From
records of the statistics department it is found that out of the first-year students who
take the basic statistics course 20% get As; out of the sophomores who take the course
30% get As; out of the juniors 35% get As; and out of the seniors who take the course
40% get As. Given that a student got an A in basic statistics, what is the probability
that she or he is a senior?
2–93. The probability that a new product will be successful if a competitor does not
come up with a similar product is 0.67. The probability that the new product will be
successful in the presence of a competitor’s new product is 0.42. The probability that
the competing firm will come out with a new product during the period in question is
0.35. What is the probability that the product will be a success?
2–94. In 2007, Starbucks inaugurated its Dulce de Leche Latte.14 If 8% of all
customers who walk in order the new drink, what is the probability that out of
13 people, at least 1 will order a Dulce de Leche Latte? What assumption are you
making?
2–95. Blackjack is a popular casino game in which the objective is to reach a card
count greater than the dealer’s without exceeding 21. One version of the game is
referred to as the “hole card” version. Here, the dealer starts by drawing a card for
himself or herself and putting it aside, face down, without the player’s seeing what it
is. This is the dealer’s hole card (and the origin of the expression “an ace in the hole”).
At the end of the game, the dealer has the option of turning this additional card face
up if it may help him or her win the game. The no-hole-card version of the game is
exactly the same, except that at the end of the game the dealer has the option of
drawing the additional card from the deck for the same purpose (assume that the
deck is shuffled prior to this draw). Conceptually, what is the difference between the
two versions of the game? Is there any practical difference between the two versions
as far as a player is concerned?
2–96. For the United States, automobile fatality statistics for the most recent
year of available data are 40,676 deaths from car crashes, out of a total population
of 280 million people. Compare the car fatality probability for one year in the
United States and in France. What is the probability of dying from a car crash in
the United States in the next 20 years?
2–97. Recall from Chapter 1 that the median is that number such that one-half the
observations lie above it and one-half the observations lie below it. If a random
sample of two items is to be drawn from some population, what is the probability that
the population median will lie between these two data points?
2–98. Extend your result from the previous problem to a general case as follows. A
random sample of n elements is to be drawn from some population and arranged
according to their value, from smallest to largest. What is the probability that the
population median will lie somewhere between the smallest and the largest values of
the drawn data?
2–99. A research journal states: “Rejection rate for submitted manuscripts: 86%.”
A prospective author believes that the editor’s statement reflects the probability of
acceptance of any author’s first submission to the journal. The author further believes
that for any subsequent submission, an author’s acceptance probability is 10% lower
than the probability he or she had for acceptance of the preceding submission. Thus,

14
Burt Helm, “Saving Starbucks’ Soul,” BusinessWeek , April 9, 2007, p. 56.
90 Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill
Complete Business Companies, 2009
Statistics, Seventh Edition

88 Chapter 2

the author believes that the probability of acceptance of a first submission to the jour-
nal is 1  0.86
0.14, the probability of acceptance of the second submission is 10%
lower, that is, (0.14)(0.90)
0.126, and so on for the third submission, fourth submi-
ssion, etc. Suppose the author plans to continue submitting papers to the journal
indefinitely until one is accepted. What is the probability that at least one paper will
eventually be accepted by the journal?15
2–100. (The Von Neumann device ) Suppose that one of two people is to be randomly
chosen, with equal probability, to attend an important meeting. One of them claims
that using a coin to make the choice is not fair because the probability that it will land
on a head or a tail is not exactly 0.50. How can the coin still be used for making the
choice? (Hint: Toss the coin twice, basing your decision on two possible outcomes.)
Explain your answer.
2–101. At the same time as new hires were taking place, many retailers were cutting
back. Out of 1,000 Kwik Save stores in Britain, 107 were to be closed. Out of 424
Somerfield stores, 424 were to be closed. Given that a store is closing, what is the
probability that it is a Kwik Save? What is the probability that a randomly chosen
store is either closing or Kwik Save? Find the probability that a randomly selected
store is not closing given that it is a Somerfield.
2–102. Major hirings in retail in Britain are as follows: 9,000 at Safeway; 5,000 at
Boots; 3,400 at Debenhams; and 1,700 at Marks and Spencer. What is the probabil-
ity that a randomly selected new hire from these was hired by Marks and Spencer?
2–103. The House Ways and Means Committee is considering lowering airline
taxes. The committee has 38 members and needs a simple majority to pass the new
legislation. If the probability that each member votes yes is 0.25, find the probability
that the legislation will pass. (Assume independence.)
Given that taxes are reduced, the probability that Northwest Airlines will com-
pete successfully is 0.7. If the resolution does not pass, Northwest cannot compete
successfully. Find the probability that Northwest can compete successfully.
2–104. Hong Kong’s Mai Po marsh is an important migratory stopover for more
than 100,000 birds per year from Siberia to Australia. Many of the bird species that
stop in the marsh are endangered, and there are no other suitable wetlands to replace
Mai Po. Currently the Chinese government is considering building a large housing
project at the marsh’s boundary, which could adversely affect the birds. Environmen-
talists estimate that if the project goes through, there will be a 60% chance that the
black-faced spoonbill (current world population
450) will not survive. It is estimat-
ed that there is a 70% chance the Chinese government will go ahead with the build-
ing project. What is the probability of the species’ survival (assuming no danger if the
project doesn’t go through)?
2–105. Three machines A, B, and C are used to produce the same part, and their
outputs are collected in a single bin. Machine A produced 26% of the parts in the bin,
machine B 38%, and machine C the rest. Of the parts produced by machine A, 8%
are defective. Similarly, 5% of the parts from B and 4% from C are defective. A part
is picked at random from the bin.
a. If the part is defective, what is the probability it was produced by machine A?
b. If the part is good, what is the probability it was produced by machine B?

15
Since its appearance in the first edition of the book, this interesting problem has been generalized. See N. H.
Josephy and A. D. Aczel, “A Note on a Journal Selection Problem,” ZOR-Methods and Models of Operations Research 34
(1990), pp. 469–76.
Aczel−Sounderpandian: 2. Probability Text © The McGraw−Hill 91
Complete Business Companies, 2009
Statistics, Seventh Edition

Probability 89

CASE
2 Job Applications

A
business graduate wants to get a job in any one should she apply to company 5? Why or why
of the top 10 accounting firms. Applying to not?
any of these companies requires a lot of effort 3. If she applies to companies 2, 5, 8, and 9, what
and paperwork and is therefore costly. She estimates is the total cost? What is the probability that she
the cost of applying to each of the 10 companies and will get at least one offer?
the probability of getting a job offer there. These data 4. If she wants to be at least 75% confident of
are tabulated below. The tabulation is in the decreas- getting at least one offer, to which companies
ing order of cost. should she apply to minimize the total cost?
(This is a trial-and-error problem.)
1. If the graduate applies to all 10 companies, what
5. If she is willing to spend $1,500, to which
is the probability that she will get at least one
companies should she apply to maximize her
offer?
chances of getting at least one job? (This is a
2. If she can apply to only one company, based trial-and-error problem.)
on cost and success probability criteria alone,

Company 1 2 3 4 5 6 7 8 9 10
Cost $870 $600 $540 $500 $400 $320 $300 $230 $200 $170
Probability 0.38 0.35 0.28 0.20 0.18 0.18 0.17 0.14 0.14 0.08