Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Chapter 4
Introduction
We are always faced with decisions and
uncertain outcomes
◦ How likely is it that the project will be
finished on time?
◦ What is the chance that a new investment will
be profitable?
◦ What is the likelihood that the Professor will
ask a question I did not study for?
Careful and precise
Probability as a Numerical Measure
of the Likelihood of Occurrence
0 1 .5
Increasing Likelihood of Occurrence
Probability:
The event
is very
________
to occur.
The occurrence
of the event is
just as likely as
it is unlikely.
The event
is almost
_________
to occur.
Experiments and Sample Spaces
Experiment: Any process that generates
well-defined outcomes
Sample Space: The set of all experimental
outcomes for an experiment
Sample Point: A particular experimental
outcome
Experiment Outcomes
Toss a coin Head, Tail
Conduct a Sales Call Purchase, No Purchase
Play a football game Win, lose, tie
Take a class A, B, C, D, F
Select product for inspection Defective, Not Defective
Example: Bradley Investments
Bradley has invested in two stocks, Markley Oil and Collins Mining. Bradley has determined that the possible outcomes of these investments three months from now are as follows.
Investment Gain or Loss
in 3 Months (in $000)
Markley Oil Collins Mining
10
5
0
-20
8
-2
Learn How to Count
Experiment: Flip a coin.
◦ How many outcomes?
Experiment: Flip a coin twice
◦ How many outcomes?
Use Tree Diagram
Counting Rule 1: If an experiment consists of a
sequence of k steps in which there are n1 possible
results for the first step, n2 possible results for the
second step, and so on, then the total number of
experimental outcomes is given by (n1)(n2) . . . (nk).
Bradley Investments can be viewed as a
two-step experiment. It involves two stocks, each with a set of experimental outcomes.
Draw a tree diagram to represent the Experimental outcomes
Markley Oil: n1 = 4
Collins Mining: n2 = 2
A Counting Rule for
Multiple-Step Experiments
Tree Diagram
Markley Oil (Stage 1)
Collins Mining (Stage 2)
Experimental Outcomes
Examples
Consider a special license plate that only has three
numbers. How many different possible license plates
are there?
______________
Delta uses a 6 letter based system for confirmation
codes. How many different combination codes are
possible with this system?
_____________________
Consider a typical license plate that uses three numbers
and three letters. How many different possible license
plates are there under this system?
Combinations and Permutations In English, we use the word „combination‟
loosely, without thinking of the order of
things
◦ Salad is a combination of lettuce, mushrooms,
and cucumbers.
◦ The combination to my locker was 5-2-7.
7-2-5 would not work!
Getting Precise
◦ Combination – Order ______ matter
◦ Permutation – Order ________ matter
Should we call this a “Permutation Lock”?
Reminder: Permutation …Position
Counting Rule 2: Combinations
n objects taken from a large set of N
objects
Objects are taken without replacement
Order does not matter.
CN
n
N
n N nnN
!
!( )!
where: N! = N(N - 1)(N - 2) . . . (2)(1)
n! = n(n - 1)(n - 2) . . . (2)(1)
0! = 1
Combination Example
An inspector randomly selects two of five parts
to test for defects. How many combinations of
two parts can be selected.
Notice that we are selecting a small set from a
larger set.
Notice that order does not matter
This is a ___________
_____________
_____________
_____________
Counting Rule 3: Permutations
n objects are to be selected from a set of
N objects and order IS important
Number of Permutations of N Objects Taken n at a Time
where: N! = N(N 1)(N 2) . . . (2)(1)
n! = n(n 1)(n 2) . . . (2)(1)
0! = 1
P nN
n
N
N nnN
!
!
( )!
•An experiment results in MORE permutations than combinations
•Every combination of n objects can be ordered in n! different ways.
Assigning Probabilities
Basic Requirements
1. The probability assigned to each experimental
outcome must be between 0 and 1
Formally: Let Ei denote the ith experimental outcome and P(Ei) its probability, then 0 ≤ P(Ei) ≤ 1 for all i
2. The sum of the probabilities must equal 1.0
Formally: For n experimental outcomes, P(E1) + P(E2) + … + P(En) = 1
Assigning Probabilities - Methods
Classical Method
◦ Assigning Probabilities based on the
assumption of Equally likely outcomes
Example
◦ Consider _____________. What probability
should we assign to each outcome?
◦ Does your method satisfy the two basic
requirements?
Assigning Probabilities - Methods Relative Frequency Method
◦ Assigning Probabilities based on experimentation
or historical data
◦ Example: Lucas Tool Rental would like to assign
probabilities to the number of car polishers it rents each
day. Office records show the following frequencies of
daily rentals for the last 40 days.
Number of Days
0 1 2 3 4
4 6 18 10 2
Number of Polishers Rented
Assigning Probabilities - Methods Relative Frequency Method
◦ Each probability assignment is given by dividing
the frequency (number of days) by the total
frequency (total number of days).
Probability Number of Polishers Rented
Number of Days
0 1 2 3 4
4 6 18 10 2 40
.10 .15 .45 .25 .05 1.00
Assigning Probabilities - Methods
Subjective Method
◦ Based on degree of belief that the
experimental outcome will occur.
◦ Use data as well as experience, intuition, but
ultimately it is our belief that matters here.
Assigning Probabilities - Methods
Subjective Method
◦ Recall Bradley‟s investment in to Oil and
Mines
Exper. Outcome Net Gain or Loss Probability
(10, 8)
(10, 2)
(5, 8)
(5, 2)
(0, 8)
(0, 2)
(20, 8)
(20, 2)
$18,000 Gain
$8,000 Gain
$13,000 Gain
$3,000 Gain
$8,000 Gain
$2,000 Loss
$12,000 Loss
$22,000 Loss
.20
.08
.16
.26
.10
.12
.02
.06
Do these satisfy the requirements for assigning probabilities?
Events and Their Probabilities Up to now: Probabilities of single sample points.
Next: Probabilities of Collections of Sample Points
Event: Collection of Sample Points
Probability of any event: Equal to the sum of the
probabilities of the sample points in the event.
If we can:
1. Identify all sample points
2. Assign probability to each sample point
Then we can compute the probability of an Event
Events and Their Probabilities
Event M = Markley Oil Profitable
(First position in the pairs below)
List the sample points in event M
What is P(M), the probability of event M? Exper. Outcome
(10, 8)
(10, 2)
(5, 8)
(5, 2)
(0, 8)
(0, 2)
(20, 8)
(20, 2)
Probability
.20
.08
.16
.26
.10
.12
.02
.06
M = {(10, 8), (10, 2), (5, 8), (5, 2)}
P(M) = P(10, 8) + P(10, 2) + P(5, 8) + P(5, 2)
= _________________
= _______________
Events and Their Probabilities
Event C = Collins Mining Profitable
(Second position in the pairs below)
List the sample points in event C
What is P(C), the probability of event C? Exper. Outcome
(10, 8)
(10, 2)
(5, 8)
(5, 2)
(0, 8)
(0, 2)
(20, 8)
(20, 2)
Probability
.20
.08
.16
.26
.10
.12
.02
.06
C = {(10, 8), (5, 8), (0, 8), (20, 8)}
P(C) = P(10, 8) + P(5, 8) + P(0, 8) + P(20, 8)
= ____________________
= ___________________
Some Basic Relationships of
Probability 1. Complement of an Event
2. Intersection of two events
3. Union of two events
4. Mutually Exclusive Events
To Understand these ideas, we are going
to use a Venn Diagram (Draw)
Some Basic Relationships of
Probability Complement of an Event
◦ The complement of Event A consists of all
sample points not in A
◦ The complement of A is denoted by Ac
Event A Ac Sample Space S
Venn Diagram
P(A) = 1 – P(Ac)
Some Basic Relationships of
Probability Intersection of two events
◦ The intersection of Events A and B is the set
of all sample points that are in _____ A and
B
The intersection of A and B is denoted
A B
A B
Some Basic Relationships of
Probability Union of two events
◦ Union of Events A and B is the event
containing all sample points that are in A
____B ____ both.
◦ The union of Events A and B is denoted by
A B. Read A _____ B
Sample Space S Event A Event B
Some Basic Relationships of
Probability Union of two events
◦ Addition Rule Enables us to calculate the
probability of two events:
◦ The probability that event A, or event B, or
both occurs is
P(AB) = P(A) + P(B) – P(A B)
A B A B = + –
Some Basic Relationships of
Probability Mutually Exclusive Events ◦ Two events are said to be mutually exclusive if
_____________________________________
◦ Two events are mutually exclusive if when one
event occurs, ______________________
Sample Space S Event A Event B
Some Basic Relationships of
Probability If events A and B are mutually exclusive,
P(A B = 0
The addition law for mutually exclusive
events is P(A B) = P(A) + P(B)
Notice, there is no need to include ____________because
________________________
A B A B =
Conditional Probability
Def: The probability of an event, given
that another event has occurred is
called conditional probability
The conditional probability of A given B is
denoted by P(A|B).
The formula is
( )
( | )( )
P A BP A B
P B
Example Consider the situation of the promotion status of male
and female officers of a major metropolitan police force
in the US. The force consists of 1200 officers, 960 men
and 420 women. Over the past two years, 324 officers
on the police force received promotions. The specific
break down is as follows.
Given someone is a man, what is the probability they
are promoted?
Given someone is a woman, what is the probability they
are promoted?
Men Women Total
Promoted 288 36 324
NotPromoted 672 204 876
Total 960 240 1200
Example Continued Process: Compute frequency information into a probability
information.
Special kind of probability information – a joint probability
distribution.
Divide each of the center table entries by the total number of
officers
Men Women Total
Promoted 288 36 324
NotPromoted 672 204 876
Total 960 240 1200
Men Women Total
Promoted .24 .03 .27
NotPromoted .56 .17 .73
Total .80 .20 1.00
•Marginal
Probabilities
because of their
location in the
margins of the
table.
•Found
_____________
_____________
__________
Example Continued
What is P(Promoted)?
_______________
What is P(Promotion | Man)?
= ______________________
= _____________
What is P(Promotion | Woman)?
= __________________________= ________________
Men Women Total
Promoted 288 36 324
NotPromoted 672 204 876
Total 960 240 1200
Men Women Total
Promoted .24 .03 .27
NotPromoted .56 .17 .73
Total .80 .20 1.00
Independent Events
Def: If the probability of Event A is not changed by the
existence of event B, we would say that events A and B
are independent
The notation for this is:
Think of police example, did gender influence
promotion?
____________________________________
If promotion was not influenced by gender, then we
could have said that promotion was independent of
gender.
P(A|B) = P(A) P(B|A) = P(B)
Multiplication Rule
Recall formula for conditional probability
We can rearrange and, viola! The
Multiplication Rule
Essentially this reminds us that if we have two
pieces of information, we can find the third.
This will come up in the homework.
( )( | )
( )
P A BP A B
P B
P(A B) = P(B)P(A|B)
Multiplication Rule – SPECIAL CASE
The Multiplication Rule
How will the Multiplication Rule change for
independent events?
Hint, independence implies
One way to test for independence is to see if
Another way is to see if
P(A B) = P(B)P(A|B)
P(A|B) = P(A)
P(A B) = P(A)P(B)
P(A|B) = P(A)
Example Suppose that we have two events, A and B. P(A) = .50, P(B) = .60, and
P(A B .40.
Find P(A | B)
Find P(B | A)
Are A and B independent? Why or why not?
Your
Turn
43 Million people in the United States go without health
insurance. Sample data representative of the national
health insurance for people older than 18 are given below
a) Develop a joint probability (including marginals) for these data and use the table.
Use the table to answer the remaining questions
b) What do the marginal age probabilities tell you about the age of the use
population 18 and older?
c) What is the probability that a randomly selected individual does not have health
insurance coverage?
d) If the individual is between 18 and 34, what is the probability that the individual
does not have health insurance coverage?
e) If the individual is 35 or older, what is the probability that the individual does not
have health insurance coverage?
f) If the individual does not have health insurance, what is the probability that the
individual is in the 18 to 34 group?
Yes No
18 to 34 750 170
35 and older 950 130
Bayes‟ Theorem
Often we begin our analysis with prior beliefs.
In probability, these are called prior
probabilities
After something happens, or we get new
information and we have to revise our
beliefs/probabilities.
Given new information, we calculate new
probabilities called Posterior Probabilities
Bayes’ Theorem is the process of getting the
revised probabilities.
Bayes‟ Theorem
Strategy we will use: Use the table familiar
from Conditional Probabilities section to
find Posterior Probabilities.
New Information
Application of Bayes’ Theorem
Posterior Probabilities
Prior Probabilities
Example
Only 1 in 1000 adults is afflicted with a rare disease for
which a diagnostic test has been developed. The test is
such that when an individual actually has the disease, a
positive result will occur 99% of the time, whereas an
individual without the disease will show a positive test
result only 2% of the time. If a randomly selected
individual is tested and the result is positive, what is the
probability that the individual has the disease?
Identify Prior information: P(disease) = ?
◦ .001
Introduce new information
◦ Someone tests positive for the disease
Form Prior: Given this new information, what is the
probability they actually have the disease?