MODERN BUSINESS STATISTICS

INTRODUCTION TO PROBABILITIES
Uncertainties
Managers often base their decisions on an analysis of uncertainties such as the following:
 What are the chances that sales will decrease if we increase prices?
 What is the likelihood a new assembly method will increase productivity?
 What are the odds that a new investment will be profitable?
Counting Rule for Combinations
Probability
 Number of Combinations of N objects taken n at a time
 Probability is a numerical measure of the likelihood that an event will occur.
A second useful counting rule enables us to count the
 Probability values are always assigned on a scale from 0 to 1. number of experimental outcomes when n objects are to be
 A probability near zero indicates an event is quite unlikely to occur. selected from a set of N objects.
 A probability near one indicates an event is almost certain to occur. Counting Rule for Permutations
Probability as a Numerical Measure of the Likelihood of Occurrence
 Number of Permutations of N Objects Taken n at a Time
A third useful counting rule enables us to count the number
of experimental outcomes when n objects are to be selected
from a set of N objects, where the order of selection is
important.
Statistical Experiments Assigning Probabilities
In statistics, the notion of an experiment differs somewhat from that of an experiment in the physical  Basic Requirements for Assigning Probabilities
sciences. In statistical experiments, probability determines outcomes. Even though the experiment is  The probability assigned to each experimental outcome must be between 0 and 1,
repeated in exactly the same way, an entirely different outcome may occur. For this reason, statistical inclusively.
experiments are sometimes called random experiments. 0 < P(Ei) < 1 for all i where: Ei is the ith experimental outcome and P(Ei) is its
An Experiment and Its Sample Space probability
 An experiment is any process that generates well-defined outcomes.  The sum of the probabilities for all experimental outcomes must equal 1.
 The sample space for an experiment is the set of all experimental outcomes. P(E1) + P(E2) + . . . + P(En) = 1 where: n is the number of experimental

 An experimental outcome is also called a sample point. outcomes

 Classical Method - assigning probabilities based on the assumption of equally likely outcomes.
Example: Rolling a Die - If an experiment has n possible outcomes, the classical method would
assign a probability of 1/n to each outcome.
 Relative Frequency Method - assigning probabilities based on experimentation or historical data.
A Counting Rule for Multiple-Step Experiments
Example: Waiting time in the X-ray department for a local
 If an experiment consists of a sequence of k steps in which there are n1 possible results for the first
hospital - Consider a study of waiting times in the X-ray
step, n2 possible results for the second step, and so on, then the total number of experimental
department for a local hospital. A clerk recorded the number of
outcomes is given by (n1)(n2) . . . (nk).
patients waiting for service at 9:00 A.M on 20 successive days
 A helpful graphical representation of a multiple-step experiment is a tree diagram.
and obtained the following results:
 Subjective Method - assigning probabilities based on judgment.
 When economic conditions or a company’s circumstances change rapidly it might be
inappropriate to assign probabilities based solely on historical data.
  We can use any data available as well as our experience and intuition, but ultimately a Example: A small assembly plant with 50 employees is carrying out performance evaluation.
probability value should express our degree of belief that the experimental outcome will Each worker is expected to complete work assignments on time and in such a way that the
occur. assembled product will pass a final inspection. The results were as follows:
 The best probability estimates often are obtained by combining the estimates from the Event L = the event that the work is completed late
classical or relative frequency approach with the subjective estimate. Event D = the event that the assembled product is
Example: Tom and Judy make an offer to purchase a house. defective
Possible outcomes: E1 = offer is accepted E2 = offer is rejected The production manager decided to assign poor
Judy’s Probability estimates P(E1) = 0.8 P(E2) = 0.2 performance rating to any employee whose work is either late or defective.
L U D = Event that the production manager assigned an employee a poor performance rating
Tom’s probability estimates P(E1) = 0.6 P(E2) = 0.4 P(L U D) = P(L) + P(D) - P(L ∩ D) P(L U D) = 0.10 + 0.12 – 0.04 = 0.18
 The assigned probabilities satisfy the two basic requirements.  Mutually Exclusive Events
 The different probability estimates emphasize the personal nature of the subjective  Two events are said to be mutually exclusive if the events have no sample points in
method. common.
Events and Their Probabilities  Two events are mutually exclusive if, when one event occurs, the other cannot occur.
 An event is a collection of sample points.  If events A and B are mutually exclusive, P(A ∩ B) = 0.
 The probability of any event is equal to the sum of the probabilities of the sample points in the  The addition law for mutually exclusive events is: P(A U B) = P(A) + P(B)
event. Conditional Probability
 If we can identify all the sample points of an experiment and assign a probability to each, we can  The probability of an event given that another event has occurred is called a conditional
compute the probability of an event. probability.
Some Basic Relationships of Probability  The conditional probability of A given B is denoted by P(A| B).
There are some basic probability relationships that can be used to compute the probability of an event Example: Promotion status of police officers over the past two years
without knowledge of all the sample point probabilities.
Men Women Total Men Women Total
 Complement of an Event Promoted 288 36 324 288/1200 = 0.24 0.03 0.27

 The complement of event A is defined to be the event consisting of all sample points that Not Promoted 672 204 876 0.56 0.17 0.73
Total 960 240 1200 0.80 0.20 1.00
are not in A.
Event A = An officer is promoted Event M = The promoted officer is a man
 The complement of A is denoted by Ac.
 P(A) + P(Ac ) = 1 P( A|M ) = An officer is promoted given that the officer is a man
 Complement of event A is shaded.
 Union of Two Events P( A ∩ M )
P ( A|M )=
 The union of events A and B is the event containing all P(M )
sample points that are in A or B or both.
From the table we know: P( A ∩ M ) = 0.24 P( M ) = 0.8 P ( A|M )=¿
 The union of events A and B is denoted by A U B.
0.24/0.8 = 0.3
 Intersection of Two Events
Multiplication Law
 The intersection of events A and B is the set of all sample points that are in both A and B.
 The multiplication law provides a way to compute the probability of the intersection of two
 The intersection of events A and B is denoted by A ∩ B. events. The law is written as: P(A ∩ B) = P(B)P(A|B)
 The addition law provides a way to compute the probability of event A, or B, or both A and B Example: Newspaper circulation department
occurring. The law is written as: P(A U B) = P(A) + P(B) - P(A ∩ B) Event D = subscribes to the daily edition Event S = subscribes to the sunday edition
 Given: P(D) = 0.84 P(S|D) = 0.75 P(S ∩ D) = P(D)P(S|D) = (0.84) (0.75) = 0.63
What is the probability that the household subscribes to both the Sunday and daily editions of the
newspaper?
Joint Probability
 Joint probabilities appear in the center of the table. From the prior probabilities and the new information we have conditional probabilities as below:
 Marginal probabilities appear in the margins of the table. P(G|A1) = .98 P(B|A1) = .02
Independent Events P(G|A2) = .95 P(B|A2) = .05

 If the probability of event A is not changed by the existence of event B, we would say that events Posterior Probability

A and B are independent.  To find the posterior probability that event Ai will occur given that event B has occurred, we apply

 Two events A and B are independent if: P(A|B) = P(A) or P(B|A) = P(B) Bayes’ theorem.

Multiplication Law for Independent Events P ( Ai ) P( B∨ A i)

 The multiplication law also can be used as a test to see if two events are independent. The law P ( A i|B )=
P ( A 1 ) P ( B| A 1) + P ( A 2 ) P ( B| A2 ) +…+ P ( A n ) P( B∨A n )
is written as: P(A ∩ B) = P(A)P(B)
Example : Use of credit card for purchase of gasoline - From past experience it is known 80% of  Bayes’ theorem is applicable when the events for which we want to compute posterior

customers use credit card for the purchase of gasoline. The service station manager wants to probabilities are mutually exclusive and their union is the entire sample space.

determine the probability that the next two customers purchasing gasoline will each use a credit card.  Example: Two supplier case

Event A: First customer uses a credit card Event B: The second customer uses a credit card Given the part received is bad we can revise the prior probabilities as below:

P(A ∩ B) = P(A)P(B) = (0.8) (0.8) = 0.64

Mutual Exclusiveness and Independence P ( A 1 ) P(B∨A 1) ( .65 )( .0 2)
 Do not confuse the notion of mutually exclusive events with that of independent events.
P ( A 1|B )= =
P ( A 1 ) P ( B| A 1) + P ( A 2) P ( B|A 2 ) ( .65 ) (. 02)+ ( .3 5 ) (.05)
 Two events with nonzero probabilities cannot be both mutually exclusive and independent.
=.4262
 If one mutually exclusive event is known to occur, the other cannot occur; thus, the probability of
In this application we started with a probability of .65 that a part selected at random was from
the other event occurring is reduced to zero (and they are therefore dependent).
supplier 1. However, given the information that the part is bad, the probability that the part is
 Two events that are not mutually exclusive, might or might not be independent.
from supplier 1 drops to .4262. In fact, if the part is bad, there is more than 50% chance that it

Bayes’ Theorem
came from supplier 2; that is, P ( A 2|B )=¿ .5738
 Often we begin probability analysis with initial or prior probabilities.
 Then, from a sample, special report, or a product test we obtain some additional information. Bayes’ Theorem: Tabular Approach
 Given this information, we calculate revised or posterior probabilities.  Step 1: Prepare the following three columns:
 Bayes’ theorem provides the means for revising the prior probabilities. Column 1 - The mutually exclusive events for which posterior probabilities are desired.
 Prior Prob. → New Info → Application of Bayes’ Theorem → Posterior Prob. Column 2 - The prior probabilities for the events.
Prior Probabilities Column 3 - The conditional probabilities of the new information given each event.
Example: Two supplier case  Step 2: Prepare the fourth column. Compute the joint probabilities for each event and the new
A manufacturing firm receives shipments of parts from two different suppliers. The quality of information B by using the multiplication law. Multiply the prior probabilities in column 2 by the
purchased parts varies with the source of supply. corresponding conditional probabilities in column 3. That is, P(Ai IB) = P(Ai) P(B|Ai).
Let: Event A1 - The part is from supplier 1 Event A2 - The part is from supplier 2  Step 3: Sum the joint probabilities in column 4.
From past experience P(A1) = .65, P(A2) = .35
Good Parts: Sup.1 – 98% Sup.2 – 95% Bad Parts: Sup.1 – 2% Sup.2 – 5%
 Step 4: In column 5, compute the posterior probabilities using the basic relationship of conditional

P( Ai ∩ B)
probability. P ( A i|B )= The joint probabilities P(A I B) are in column 4 and
i
P (B)
the probability P(B) is the sum of column 4.

 Inference from the table: There is a .0130 probability that the part came from supplier 1 and is
bad. There is .0175 probability that the part
came from supplier 2 and is bad. The overall
probability of finding a bad part from the
combined shipments of two suppliers is .0305.