PROBABILITY

Managers often base their decisions on an

analysis of Uncertainties such as
following:
What are chances that sales will decrease if
we increase prices?
How likely is it that the project will finish on
time?
What is the chance that a new investment will
be profitable?
Definition:

Probability is a numerical measure of the
likelihood that an event will occur.

Probabilities values are always assigned on a
scale from 0-1.
A probability near zero indicates an event is
unlikely to occur
A probability near one indicates an event is
likely to occur
Other probabilities between 0 & 1 represents
degrees of likelihood that an event will
occur.
EXPERIMENT
An experiment is a process that generates
well-defined outcomes.

On any single repetition of an experiment,
one and only one of the possible experimental
outcomes will occur.
Example
EXPERIMENT EXPERIMENTAL
OUTCOME
Toss a coin Head, Tail
Select a part for inspection Defective, Non-Defective
Conduct a sales call Purchase, No Purchase
Roll a die 1, 2, 3, 4, 5, 6
Play a football game Win, Lose, Tie
Sample Space
The sample space for an experiment is the
set of all experimental outcomes.

Sample Points: Experimental outcomes are
also called sample points.
Sample Space for the first
experiment in the previous table
Experiment 1
S = {H, T}

Experiment 2
S = {Defective, Non Defective}

Experiment 4
S = {1,2, 3, 4, 5, 6 }
Example 1: Gender of Two
Children
A newly married couple plan to have two
children. Naturally, they are curious about
whether their children will be boys or girls.
Therefore, we consider the experiment of
having two children.
In order to find the sample space let
B denote the child is a boy
G denote the child is a girl
Therefore, the sample space is
S = { BB , BG , GB , GG }
In order to consider the probabilities of these
outcomes, suppose that boys and girls are
equally likely each time a child is born.
Intuitively, this says that each of the sample
space outcomes is equally likely.


This says that there is a 25 % chance of these
outcomes will occur. These probabilities sum to
1.

1
( ) ( ) ( ) ( )
4
P BB P BG P GB P GG = = = =
A Tree Diagram of the Genders of
Two Children
Events
An event is a set (or collection) of sample
space outcomes
The probability of an event is the sum of the
probabilities of the sample space outcomes
that correspond to the event
Example 1: Gender of Two Children
Experimental Outcomes:
BB, BG, GB, GG
All outcomes equally likely:
P(BB) = = P(GG) =
P(one boy and one girl) =
P(BG) + P(GB) = + =
P(at least one girl) =
P(BG) + P(GB) + P(GG) = ++ =
Probabilities: Equally Likely
Outcomes
If the sample space outcomes (or
experimental outcomes) are all equally
likely, then the probability that an event will
occur is equal to the ratio:
The number of ways the event can occur
Over the total number of outcomes
outcomes space sample of number Total
event the to correspond that outcomes space sample of Number
Example 2: Newspapers Subscribers
Suppose that 650,000 of the 1,000,000 households in an
eastern U.S. city subscribe to a newspaper called the
Atlantic Journal, and consider randomly selecting one of
the households in this city. That is, consider selecting
one household by giving each and every household in
the city the same chance of being selected.
Let A be the event that the randomly selected
households subscribes to the Atlantic Journal. Then
because the sample space of this experiment consists of
1,000,000 equally likely sample space outcomes
(households), it follows that
the number of households that subscribes to the Atlantic Journal
the total number of households in the city
650, 000
1, 000, 000
.65
=
=
This says that the probability that the randomly selected
Household subscribes to the Atlantic Journal is .65
Counting Rules
Being able to identify and count the
experimental outcomes is a necessary step in
assigning probabilities.

Multiple Step Experiments
Consider the experiment of tossing two coins.
How many experimental outcomes are possible for
this experiment?

Step 1: Tossing of the first coin.
Step 2: Tossing of the second coin.

Therefore,
S = {(H, H), (H, T), (T, H), (T, T)}
Counting Rule for multiple
step experiments
If an experiment can be described as a
sequence of k steps with n1, possible
outcomes on the first step, n2 possible
outcomes on the second step, and so on,
then the total number of experimental
outcomes is given by (n1).(n2)(nk)
Counting Rule for:
Tossing two coins
(2) (2) = 4
Tossing three coins
(2) (2) (2) = 8
Rolling two die
(6) (6) = 36

Case:
How the counting rule for multiple step experiments
can be used in the analysis of a capacity expansion
project for the Kentucky Power and Light Company
(KP&L). KP&L is starting a project designed to
increase the generating capacity of one of its plants
in Northern Kentucky. The project is divided into
sequential stages
Stage 1: Design
Stage 2: Construction
An analysis of similar construction projects
reveled possible completion times for the
design stage of 2, 3, or 4 months and
possible completion times for the construction
stage of 6, 7, or 8 months.

Because of critical need for additional
electrical power , management set a goal of
10 months for the completion of the entire
project
Experimental Outcomes for KP&L Project
Stage 1
(Design)
Stage 2
(Construction)
Notation for
Experimental
Outcome
Total Project
Completion
time
2 6 (2, 6) 8
2 7 (2, 7) 9
2 8 (2, 8) 10
3 6 (3, 6) 9
3 7 (3, 7) 10
3 8 (3, 8) 11
4 6 (4, 6) 10
4 7 (4, 7) 11
4 8 (4, 8) 12
Assigning Probabilities
Classical Method
Relative Frequency Method
Subjective Method
Basic Requirements for
Assigning Probabilities
The probability assigned to each experimental
outcome must be between 0 and 1


The sum of the probabilities for all the
experimental outcomes must equal 1.0. For n
experimental outcomes, this requirement can be
written as
| | i all for E P
i
1 0 s s
| | | | | | 1 .. ..........
2 1
= + + +
n
E P E P E P
Classical Method
The classical method of assigning
probabilities is appropriate when all the
experimental outcomes are equally likely.
If n experimental outcomes are possible, a
probability of 1/n is assigned to each
experimental outcome.
Example:
Tossing of a fair coin.
Rolling a die

The Relative Frequency
Method
Appropriate when data are available to
estimate the proportion of the time the
experimental outcome will occur if the
experiment is repeated to large number of
times.
Example: Consider a study of waiting times in
the X-ray department for a local hospital. A
clerk recorded the number of patients waiting
for service at 9:00 a.m. on 20 successive
days obtained the following results.
Number Waiting Number of Days
Outcome Occurred
0 2
1 5
2 6
3 4
4 3
Total 20
Subjective Method
Appropriate when one cannot realistically
assume that the experimental outcomes are
equally likely and when little relevant data are
available.
Any available information may be used such
as our experience or intuition.

Example:
Consider the case in which Tom and Jack make an
offer to purchase a house.
Two possible outcomes are:
= their offer is accepted.
= their offer is rejected.
If Jack believes that the probability their offer will be
Accepted is 0.8.
Thus

1
E
2
E
| |
| | 2 . 0
8 . 0
2
1
=
=
E P
E P
Event and their Probabilities
An event is a collection of sample points.
Let C denote the event that the project is
completed in ten months or less.
C = {(2, 6), (2, 7), (2, 8), (3, 6), (3, 7), (4, 6)}
L = The event that the project is completed in
less than 10 months
{(2, 6), (2, 7), (3,6)}

Stage 1
(Design)
Stage 2
(Construction)
Sample
Point
No. of Past Projects
having these
Completion times
2 6 (2, 6) 6
2 7 (2, 7) 6
2 8 (2, 8) 2
3 6 (3, 6) 4
3 7 (3, 7) 8
3 8 (3, 8) 2
4 6 (4, 6) 2
4 7 (4, 7) 4
4 8 (4, 8) 6
Completion Time (months)
Total = 40
Sample Point Project
Completion
Time
Probability of
Sample Point
(2, 6) 8 months P(2, 6) = 6/40 =.15
(2, 7) 9 months P(2, 7) = 6/40 =.15
(2, 8) 10 months P(2, 8) = 2/40 =.05
(3, 6) 9 months P(3, 6) = 4/40 =.10
(3, 7) 10 months P(3, 7) = 6/40 =.20
(3, 8) 11 months P(3, 8) = 6/40 =.05
(4, 6) 10 months P(4, 6) = 6/40 =.05
(4, 7) 11 months P(4, 7) = 6/40 =.10
(4, 8) 12 months P(4, 8) = 6/40 =.15
Total = 1.00
Probability of event C (i.e. project is completed in 10
months or less) is given by:


Similarly, probability that the project is completed in
less than 10 months is given by:

( ) (2, 6) (2, 7) (2,8) (3, 6) (3, 7) (4, 6)
= 0.15 0.15 0.05 0.10 0.20 0.05 0.70
P C P P P P P P = + + + + +
+ + + + + =
( ) (2, 6) (2, 7) (3, 6)
= 0.15 0.15 0.10 0.40
P L P P P = + +
+ + =
Complement of an Event
Given an event A, the complement of A is
defined to be the event consisting of all sample
points that are not in A. The complement of A is
denoted by
P(A) + P( ) = 1
c
A
c
A
Event A
c
A
Sample Space S
Complement of Event A
Computing Probability Using the
Complement

P(A) = 1 P( )
Example:
Consider the case of a sales manager who, after
reviewing sales reports, states that 80% of new
customer contacts result in no sale.

By allowing A = the event of a sale and
= the event of no sale
Using the above eqn
P(A) = .20

c
A
c
A
Example 2:
The probability that a randomly selected household
In an eastern U.S. city subscribers to the Atlantic Journal is
.65.
It follows that the probability of the complement of this event
Is 1 - .65 = .35

Addition Law
The addition law is helpful when we are
interested in knowing the probability that at
least one of the two occurs.
That is, with events A and B we are
interested in knowing the probability that
event A or event B or both occur.
Union of two Events
The union of A and B is the event containing
all sample points belonging to A or B or both.
The union is denoted by A U B.
Sample Space S
Event A Event B
Intersection of Two Events
Given two events A and B, the intersection of A and
B is the event containing the sample points
belonging to both A and B. The intersection is
denoted by A B.



ADDITION LAW
P(A U B) = P(A) + P(B) P(A B )

Example:
There is a small assembly plant with 50 employees .
Each worker is expected to complete work
assignments on time and in such a way that the
assembled product will pass a final inspection. On
occasion, some of the workers failed to meet the
performance standards by completing work late
or assembling a defective product. At the end of a
performance evaluation period, the production
manager found that 5 of 50 workers completed work
late, 6 of 50 workers assembled a defective product,
and 2 of the 50 workers both completed work late
and assembled a defective product .
Let
L = the event that the work is completed late
D = the event that the assembled product is
defective
The relative frequency information leads to the following
probabilities







After reviewing the performance data, the production manager
decided to assign a poor performance rating to any employee whose
work was either late or defective; thus the event of interest is L U D.

50
04 . 2 D) P(L
50
12 . 6 P(D)
50
10 . 5 ) (
= =
= =
= = L P
What is the probability that the production
manager assigned an employee a poor
performance rating?
P(L U D) = P(L) + P(D) P(L D)
P(L U D) = .10 + .12 - .04 = .18

This calculation tells us that there.18 is a probability
that a randomly selected employee received a poor
performance rating.

Example 2: Newspaper Subscribers

Define events:
A = event that a randomly selected household
subscribes to the Atlantic Journal
B = event that a randomly selected household
subscribes to the Beacon News
Given:
total number in city, N = 1,000,000
number subscribing to A, N(A) = 650,000
number subscribing to B, N(B) = 500,000
number subscribing to both, N(AB) = 250,000
Example: Newspaper Subscribers
Use the relative frequency method to assign
probabilities
( )
( )
( ) 25 . 0
000 , 000 , 1
000 , 250
50 . 0
000 , 000 , 1
000 , 500
65 . 0
000 , 000 , 1
000 , 650
= =
= =
= =
B A P
B P
A P

Define events:
A = event that a randomly selected household
subscribes to the Atlantic Journal.
= event that a randomly selected household does
not subscribes to the Atlantic Journal.
B = event that a randomly selected household
subscribes to the Beacon News.
= event that a randomly selected household does
not subscribes to the Beacon News.

A
B
Table 2.1: A Summary of the Number of
Households Corresponding to the Events
A, , B, and AB
Events Subscribes to
Beccon News, B
Does Not
subscribe to
Beacon News,
Total
Subscribes to
Atlantic Journal, A
2,50,000 6,50,000
Does Not
Subscribes to
Atlantic Journal,
3,50,000
Total 5,00,000 5,00,000 10,00,000
B
A
A B
Table 2.2: The number of Households
corresponding to
( and ) A B
Events Subscribes to
Beccon News, B
Does Not subscribe to
Beacon News,
Total
Subscribes to
Atlantic Journal, A
2,50,000 6,50,000 2,50,000 =
4,00,000
6,50,000
Does Not
Subscribes to
Atlantic Journal,
3,50,000
Total 5,00,000 5,00,000 10,00,000
B
A
Table 2.3: The number of Households
corresponding to
( and B) A
Events Subscribes to
Beccon News, B
Does Not subscribe
to Beacon News,
Total
Subscribes to
Atlantic Journal, A
2,50,000 6,50,000 2,50,000 =
4,00,000
6,50,000
Does Not
Subscribes to
Atlantic Journal,
5,00,000 2,50,000 =
2,50,000
3,50,000
Total 5,00,000 5,00,000 10,00,000
A
B
Table 2.4: The number of Households
corresponding to
( and B) A
Events Subscribes to
Beccon News, B
Does Not subscribe
to Beacon News,
Total
Subscribes to
Atlantic Journal, A
2,50,000 6,50,000 2,50,000 =
4,00,000
6,50,000
Does Not
Subscribes to
Atlantic Journal,
5,00,000 2,50,000 =
2,50,000
3,50,000 2,50,000 =
1,00,000
3,50,000
Total 5,00,000 5,00,000 10,00,000
A
B
Table 2.5: A Contingency Table
summarizing Subscription Data for the
Atlantic Journal and the Beacon News
Events Subscribes to
Beccon News, B
Does Not subscribe
to Beacon News,
Total
Subscribes to
Atlantic Journal, A
2,50,000 4,00,000 6,50,000
Does Not
Subscribes to
Atlantic Journal,
2,50,000 1,00,000 3,50,000
Total 5,00,000 5,00,000 10,00,000
A
B
Example: Newspaper Subscribers
Refer to the contingency table in Table 2.5 for
all data
For example, the chance that a household
does not subscribe to either newspaper
Want , so from middle row and middle
column of Table 2.5
( ) 10 0
000 000 1
000 100
B A P .
, ,
,
= =
( ) B A P
Example: Newspaper Subscribers
The chance that a household subscribes to
either newspaper:



Note that if the joint probability was not
subtracted it would have gotten 1.15, which is
greater than 1, which is absurd
The subtraction avoids double counting the joint
probability
90 . 0
25 . 0 50 . 0 65 . 0
=
+ =
B) P(A B) B)=P(A)+P( P(A
Conditional Probability
Often, the probability of an event is
influenced by whether related event already
occurred.
P(A / B)
Indicate that we are considering the
probability of event A given the condition that
event B has occurred.
Example
Consider the situation of the promotion status of male
and female officers of a major metropolitan police force
in eastern united states. The police force consists of
1200 officers, 960 men and 240 women. Over the past
years , 324 officers on the police force received
promotions. After reviewing the promotion record, a
committee of female officers raised a discrimination case
on the basis that 288 male officers had received
promotions but only 36 female officers had been
promoted. The police administration argued that the
relatively low number of promotions for female officers
was due not to discrimination, but to the fact that
relatively few females are members of police force.
Let
M = event an officer is a man.
W = event an officer is a woman.
A = event an officer is promoted.
A = event an officer is not promoted.

P(M A) = 288/1200 = .24 = prob. that a randomly
selected officer is a man and is promoted.
P(M A') = 672/1200 = .56 = prob. that a randomly
selected officer is a man and is not promoted.
P(W A) = 36/1200 = prob. that a randomly
selected officer is a woman and is promoted.
P(W A') = 204/1200 = prob. that a randomly
selected officer is a woman and is not promoted.

Because each of these values gives the probability of the
intersection of two events, the probabilities are called joint
probabilities.
Promotion Status of Police officers Over the
Past Two Years
Men Women Totals
Promoted 288 36 324
Not Promoted 672 204 876
Totals 960 240 1200
Joint Probability Table for Promotions


Men (M) Women (W) Totals
Promoted (A) .24 .03 .27
Not Promoted (A') .56 .17 .73
Totals .80 .20 1.00
Joint probabilities
appear in the body
of the table.
Marginal probabilities
appear in the
margins of the table.
Conditional Probability
Therefore, conditional probability P(A/M) can be computed as
the ratio of the joint probability P(AM) to the marginal
probability P(M).

P(A / M) = P(AM) = .24 = .30
P(M) .80

P(A / W) = P(AW) = .03 = .15
P(W) .20

Thus, the conditional probability values support the argument
presented by the female officers.

Independent Events
P(A) = .27, P(A / M) = .30 and P(A / W) = .15
Probability of a promotion (event A) is
affected or influenced by whether the officer
is a man or women.
Particularly because P(A / M) P(A), we
would say that prob. A and M are dependent
events.
Two events A & B are
independent if,
P(A / B) = P(A)

P(B / A) = P(B)

Otherwise the events are dependent.
Multiplication Law
It computes the probability of two events.
It is based on the definition of conditional
probability

P(AB) = P(B) . P(A/B)

Or P(AB) = P(A) . P(B/A)
Example:
Consider a newspaper circulation department where it is
known that 84% of the households in a particular
neighborhood subscribe to the daily edition of the paper.
If we let D denote the event that a household subscribes
to the daily edition P(D) = .84. In addition, it is known that
the probability that a household that already holds a daily
subscription also subscribes to the Sunday edition (event
S) is .75 i.e. P(S / D) = .75. What is the probability that a
household subscribes to both the Sunday and daily
editions of the newspaper?

Solution:
P(SD) = P(D) . P(S / D)
= .84 .75 = .63

Therefore 63% of the households subscribes
to both Sunday and daily edition.
Multiplication Law for
Independent Events
P(AB) = P(A) . P(B)

Example: A boy wants to marry a girl having
qualities: white complexion, the probability of getting
such a girl is one in twenty; handsome dowry the
probability of getting such a girl is one in fifty;
Westernized manners and etiquettes, the probability
here is one in hundred. Find out the probability of
his getting married to such a girl when the
possession of these three attributes is independent.
Solution:
Prob. Of a girl with white complexion
P(W) = 1/20 = 0.05
Prob. Of a girl with handsome dowry
P(D) = 1/50 = 0.02
Prob. Of a girl with westernized manners & etiquettes
P(E) = 1/100 = 0.01
Since the events are independent, the probability of
simultaneous occurrence of all these qualities.
P(M) = P(W) . P(D) . P(E)
= 0.05 . 0.02 . 0.01 = 0.00001
= 1/ 100000


Bayes Theorem
In the discussion of conditional probability, we indicated
that revising probability when new information is
obtained is an important phase of probability analysis.
Often analysis is begin with initial or prior probability
estimates for specific events of interest.
Then, from sources such as a sample, a special report,
or a product test, we obtain additional information about
the events.
Given this new information, prior probability values gets
calculated and revised, referred to as posterior
probabilities.
Bayes theorem provides a means for making these
probability calculations.
Prior
Probabilities
New
Information
Application
Of Bayes
Theorem
Posterior
Probabilities
Formula



) / ( ) (
n) ....., 2, 1, .....(i ) / ( ) ( ) / (
i i
i i i
B A P B P
B A P B P A B P

= =
occured. has A that
n informatio additional the of because ies probabilit priori' a '
in change indicates that ies' probabilit posterior ' called are
/A)s P(B and ies, probabilit priori' a ' called are )s P(B Here,
i i
Example
Three machines producing 40%, 35%, and
25% of the total output are known to produce
with defective proportion of items as: 0.04,
0.06, and 0.03 respectively. On a particular
day, a unit of output is selected at random,
and is found to be defective. What is the
probability that it was produced by the
second machine?
Solution
machines three the of either from come to has item an i.e.
1 ) P(M ) P(M ) P(M
that noted be to is It
0.25 ) P(M
0.35 ) P(M
0.40 ) P(M
given that is It
. M machine by produced being item an of ies Probabilit ) P(M
. M machine by produced being item an of ies Probabilit ) P(M
. M machine by produced being item an of ies Probabilit ) P(M
M and , M , M as labeled be machines three Let the
3 2 1
3
2
1
3 3
2 2
1 1
3 2 1
= + +
=
=
=
=
=
=
0.472
.03 .25 .06 .35 .04 .40
0.06 0.35
) / ( ) P(M
) P(D/M ) P(M /D) P(M
M machine by produced it was that prob. the is what defective,
being item the i.e. /D) P(M y probabilit the find to required are We
0.03 ) P(D/M
0.06 ) P(D/M
0.04 ) P(D/M
. M , M , M machines by the produced being item defective of
ies probabilit the denote ) P(D/M ), P(D/M ), P(D/M let Further,
i
2 2 2
2
2
3
2
1
3 2 1
3 2 1
=
+ +
=
=
=
=
=
i
M D P