PPT Render Ch 2
-
Upload
jenine-quiambao -
Category
Documents
-
view
223 -
download
0
Transcript of PPT Render Ch 2
-
7/28/2019 PPT Render Ch 2
1/91
2008 Prentice-Hall, Inc.
Chapter 2
To accompanyQuant i tat ive Analysis for Management, Tenth Edit io n,
by Render, Stair, and HannaPower Point slides created by Jeff Heyl
Probab i li ty Concep ts and
Appl icat ions
2009 Prentice-Hall, Inc.
-
7/28/2019 PPT Render Ch 2
2/91
2008 Prentice-Hall, Inc. 2 2
Learn ing Object ives
1. Understand the basic foundations ofprobability analysis
2. Describe statistically dependent andindependent events
3. Use Bayes theorem to establish posteriorprobabilities
4. Describe and provide examples of both
discrete and continuous random variables5. Explain the difference between discrete and
continuous probability distributions
6. Calculate expected values and variances
and use the normal table
After completing this chapter, students will be able to:
-
7/28/2019 PPT Render Ch 2
3/91
2008 Prentice-Hall, Inc. 2 3
Chapter Ou tl ine
2.1 Introduction
2.2 Fundamental Concepts
2.3 Mutually Exclusive and Collectively
Exhaustive Events2.4 Statistically Independent Events
2.5 Statistically Dependent Events
2.6 Revising Probabilities with BayesTheorem
2.7 Further Probability Revisions
-
7/28/2019 PPT Render Ch 2
4/91
2008 Prentice-Hall, Inc. 2 4
Chapter Ou tl ine
2.8 Random Variables
2.9 Probability Distributions
2.10 The Binomial Distribution
2.11 The Normal Distribution2.12 The FDistribution
2.13 The Exponential Distribution
2.14 The Poisson Distribution
-
7/28/2019 PPT Render Ch 2
5/91
2008 Prentice-Hall, Inc. 2 5
In t roduct ion
Life is uncertain, we are not surewhat the future will bring
Risk and probability is a part ofour daily lives
Probabi l i tyis a numerical
statement about the likelihoodthat an event will occur
-
7/28/2019 PPT Render Ch 2
6/91
2008 Prentice-Hall, Inc. 2 6
Fundamental Concepts
1. The probability, P, of any event orstate of nature occurring is greaterthan or equal to 0 and less than or
equal to 1. That is:
0 P(event) 1
2. The sum of the simple probabilitiesfor all possible outcomes of anactivity must equal 1
-
7/28/2019 PPT Render Ch 2
7/91 2008 Prentice-Hall, Inc. 2 7
Chapters in This BookThat Use Probabi l i ty
CHAPTER TITLE
3 Decision Analysis
4 Regression Models
5 Forecasting
6 Inventory Control Models
13 Project Management
14 Waiting Lines and Queuing Theory Models
15 Simulation Modeling
16 Markov Analysis17 Statistical Quality Control
Module 3 Decision Theory and the Normal Distribution
Module 4 Game Theory
Table 2.1
-
7/28/2019 PPT Render Ch 2
8/91 2008 Prentice-Hall, Inc. 2 8
Diversey Paint Examp le
Demand for white latex paint at Diversey Paintand Supply has always been either 0, 1, 2, 3, or 4gallons per day
Over the past 200 days, the owner has observed
the following frequencies of demandQUANTITY
DEMANDEDNUMBER OF DAYS PROBABILITY
0 40 0.20 (= 40/200)
1 80 0.40 (= 80/200)
2 50 0.25 (= 50/200)
3 20 0.10 (= 20/200)
4 10 0.05 (= 10/200)
Total 200 Total 1.00 (= 200/200)
-
7/28/2019 PPT Render Ch 2
9/91 2008 Prentice-Hall, Inc. 2 9
Diversey Paint Examp le
Demand for white latex paint at Diversey Paintand Supply has always been either 0, 1, 2, 3, or 4gallons per day
Over the past 200 days, the owner has observed
the following frequencies of demandQUANTITY
DEMANDEDNUMBER OF DAYS PROBABILITY
0 40 0.20 (= 40/200)
1 80 0.40 (= 80/200)
2 50 0.25 (= 50/200)
3 20 0.10 (= 20/200)
4 10 0.05 (= 10/200)
Total 200 Total 1.00 (= 200/200)
Notice the individual probabilitiesare all between 0 and 1
0 P(event) 1
And the total of all eventprobabilities equals 1
P(event) = 1.00
-
7/28/2019 PPT Render Ch 2
10/91 2008 Prentice-Hall, Inc. 2 10
Determining ob ject ive probabi l i ty
Relative frequency Typically based on historical data
Types o f Probab i l ity
P(event) =Number of occurrences of the event
Total number of trials or outcomes
Classical or logical method
Logically determine probabilities withouttrials
P(head) =1
2
Number of ways of getting a head
Number of possible outcomes (head or tail)
-
7/28/2019 PPT Render Ch 2
11/91 2008 Prentice-Hall, Inc. 2 11
Types o f Probabi l i ty
Subject ive probabi l i tyis based onthe experience and judgment of theperson making the estimate
Opinion polls
Judgment of experts
Delphi method
Other methods
-
7/28/2019 PPT Render Ch 2
12/91 2008 Prentice-Hall, Inc. 2 12
Mutual ly Exc lus ive Even ts
Events are said to be mutual lyexclus iveif only one of the events canoccur on any one trial
Tossing a coin will resultin eithera head or a tail
Rolling a die will result inonly one of six possibleoutcomes
-
7/28/2019 PPT Render Ch 2
13/91 2008 Prentice-Hall, Inc. 2 13
Col lect ively Exhaus t ive Events
Events are said to be collectivelyexhaustive if the list of outcomesincludes every possible outcome
Both heads andtails as possibleoutcomes ofcoin flips
All six possibleoutcomesof the rollof a die
OUTCOMEOF ROLL
PROBABILITY
1 1/6
2 1/6
31
/64 1/6
5 1/6
6 1/6
Total 1
-
7/28/2019 PPT Render Ch 2
14/91 2008 Prentice-Hall, Inc. 2 14
Draw ing a Card
Draw one card from a deck of 52 playing cards
P(drawing a 7) = 4/52 =1/13
P(drawing a heart) = 13/52 = 1/4
These two events are not mutually exclusivesince a 7 of hearts can be drawn
These two events are not collectivelyexhaustive since there are other cards in thedeck besides 7s and hearts
-
7/28/2019 PPT Render Ch 2
15/91
2008 Prentice-Hall, Inc. 2 15
Tab le o f Differences
DRAWSMUTUALLYEXCLUSIVE
COLLECTIVELYEXHAUSTIVE
1. Draws a spade and a club Yes No
2. Draw a face card and anumber card Yes Yes
3. Draw an ace and a 3 Yes No
4. Draw a club and a nonclub Yes Yes
5. Draw a 5 and a diamond No No
6. Draw a red card and adiamond
No No
-
7/28/2019 PPT Render Ch 2
16/91
2008 Prentice-Hall, Inc. 2 16
Add ing Mutually Exc lusive Events
We often want to know whether one or asecond event will occur
When two events are mutually
exclusive, the law of addition is
P(event A or event B) = P(event A) + P(event B)
P(spade or club) = P(spade) + P(club)
= 13/52 +13/52
= 26/52 =1/2 = 0.50 = 50%
-
7/28/2019 PPT Render Ch 2
17/91
2008 Prentice-Hall, Inc. 2 17
Add ing Not Mutually Exclus ive Events
P(event A or event B) = P(event A) + P(event B)
P(event Aand event Bbothoccurring)
P(A orB) = P(A) + P(B)P(A and B)
P(five or diamond) =P(five) + P(diamond)P(five and diamond)
= 4/52 +13/52
1/52= 16/52 =
4/13
The equation must be modified to accountfor double counting
The probability is reduced bysubtracting the chance of both eventsoccurring together
-
7/28/2019 PPT Render Ch 2
18/91
2008 Prentice-Hall, Inc. 2 18
Venn Diagrams
P(A) P(B)
Events that are mutuallyexclusive
P(A orB) = P(A) + P(B)
Figure 2.1
Events that are notmutually exclusive
P(A orB) = P(A) + P(B)P(A and B)
Figure 2.2
P(A) P(B)
P(A and B)
-
7/28/2019 PPT Render Ch 2
19/91
2008 Prentice-Hall, Inc. 2 19
Stat ist ical ly Independen t Even ts
Events may be either independent ordependent For independent events, the occurrence
of one event has no effect on theprobability of occurrence of the secondevent
-
7/28/2019 PPT Render Ch 2
20/91
2008 Prentice-Hall, Inc. 2 20
Which Sets o f Events A re Independent?
1. (a) Your education
(b) Your income level
2. (a) Draw a jack of hearts from a full 52-card deck
(b) Draw a jack of clubs from a full 52-card deck
3. (a) Chicago Cubs win the National League pennant
(b) Chicago Cubs win the World Series
4. (a) Snow in Santiago, Chile
(b) Rain in Tel Aviv, Israel
Dependent events
Dependent
events
Independent
events
Ind ependent events
-
7/28/2019 PPT Render Ch 2
21/91
2008 Prentice-Hall, Inc. 2 21
Three Types o f Probab i l i t ies
Marginal(ors imple) probability is just theprobability of an event occurring
P(A)
Jo in tprobability is the probability of two or more
events occurring and is the product of theirmarginal probabilities for independent events
P(AB) = P(A) x P(B)
Condi t ionalprobability is the probability of event
Bgiven that event A has occurredP(B| A) = P (B)
Or the probability of event A given that event Bhas occurred
P(A | B) = P(A)
-
7/28/2019 PPT Render Ch 2
22/91
2008 Prentice-Hall, Inc. 2 22
Jo int Probabi l i ty Example
The probability of tossing a 6 on the firstroll of the die and a 2 on the second roll
P(6 on first and 2 on second)
= P(tossing a 6) x P(tossing a 2)
= 1/6 x1/6 =
1/36 = 0.028
-
7/28/2019 PPT Render Ch 2
23/91
2008 Prentice-Hall, Inc. 2 23
Independen t Even ts
1. A black ball drawn on first drawP(B) = 0.30 (a marginal prob abi l ity)
2. Two green balls drawn
P(GG) = P(G) x P(G) = 0.7 x 0.7 = 0.49
(a joint p robabi l i ty for two ind ependent events)
A bucket contains 3 black balls and 7 green balls We draw a ball from the bucket, replace it, and
draw a second ball
-
7/28/2019 PPT Render Ch 2
24/91
2008 Prentice-Hall, Inc. 2 24
Independen t Even ts
3. A black ball drawn on second draw if the firstdraw is green
P(B|G) = P(B) = 0.30(a cond it ional probabi l i ty but equal to the marginalbecause the two d raws are independent events)
4. A green ball is drawn on the second if the firstdraw was green
P(G|G) = P(G) = 0.70(a cond it ional probabi l i ty as in event 3)
A bucket contains 3 black balls and 7 green balls We draw a ball from the bucket, replace it, and
draw a second ball
-
7/28/2019 PPT Render Ch 2
25/91
2008 Prentice-Hall, Inc. 2 25
Statist ical ly Dependen t Events
The marginalprobability of an event occurring iscomputed the same
P(A)
The formula for thejo in tprobability of two events is
P(AB) = P(B|A) P(A)
P(A | B) =P(AB)
P(B)
Calculating condi t ionalprobabilities is slightly morecomplicated. The probability of event A given thatevent Bhas occurred is
-
7/28/2019 PPT Render Ch 2
26/91
2008 Prentice-Hall, Inc. 2 26
When Events Are Dependent
Assume that we have an urn containing 10 balls ofthe following descriptions
4 are white (W) and lettered (L )
2 are white (W
) and numbered (N
)
3 are yellow (Y) and lettered (L )
1 is yellow (Y) and numbered (N)
P(WL
) = 4/10
= 0.4P
(YL
) = 3/10
= 0.3
P(WN) = 2/10 = 0.2 P(YN) =1/10 = 0.1
P(W) = 6/10 = 0.6 P(L ) =7/10 = 0.7
P(Y) = 4/10 = 0.4 P(N) =3/10 = 0.3
-
7/28/2019 PPT Render Ch 2
27/91
2008 Prentice-Hall, Inc. 2 27
When Events Are Dependent
4 ballsWhite (W)
andLettered (L )
2 ballsWhite (W)
andNumbered (N)
3 ballsYellow (Y)
andLettered (L )
1 ball Yellow (Y)and Numbered (
N)
Probability (WL ) =4
10
Probability (YN) =1
10
Probability (YL ) =310
Probability (WN) =210
Urn contains10 balls
Figure 2.3
-
7/28/2019 PPT Render Ch 2
28/91
2008 Prentice-Hall, Inc. 2 28
When Events Are Dependent
The conditional probability that the ball drawnis lettered, given that it is yellow, is
P(L| Y) = = = 0.75P(YL )
P(Y)
0.3
0.4
Verify P(YL ) using the joint probability formula
P(YL ) = P(L| Y) x P(Y) = (0.75)(0.4) = 0.3
-
7/28/2019 PPT Render Ch 2
29/91
2008 Prentice-Hall, Inc. 2 29
Jo int Probabi l i t iesfor Dependent Events
P(MT) = P(T| M) x P(M) = (0.70)(0.40) = 0.28
If the stock market reaches 12,500 point by January,there is a 70% probability that Tubeless Electronicswill go up
There is a 40% chance the stock market will
reach 12,500 Let Mrepresent the event of the stock market
reaching 12,500 and let Tbe the event thatTubeless goes up in value
-
7/28/2019 PPT Render Ch 2
30/91
2008 Prentice-Hall, Inc. 2 30
PosteriorProbabilities
BayesProcess
Revis ing Probabi l i t ies w ith
Bayes Theorem
Bayes theorem is used to incorporate additionalinformation and help create pos ter ior prob abi l i t ies
PriorProbabilities
NewInformation
Figure 2.4
-
7/28/2019 PPT Render Ch 2
31/91
2008 Prentice-Hall, Inc. 2 31
Pos terio r Probab i l i t ies
A cup contains two dice identical in appearance butone is fair (unbiased), the other is loaded (biased)
The probability of rolling a 3 on the fair die is 1/6 or 0.166
The probability of tossing the same number on the loadeddie is 0.60
We select one by chance,toss it, and get a result of a 3
What is the probability thatthe die rolled was fair?
What is the probability that
the loaded die was rolled?
-
7/28/2019 PPT Render Ch 2
32/91
2008 Prentice-Hall, Inc. 2 32
Pos terio r Probab i l i t ies
We know the probability of the die being fair orloaded is
P(fair) = 0.50 P(loaded) = 0.50
And that
P(3 | fair) = 0.166 P(3 | loaded) = 0.60
We compute the probabilities ofP(3 and fair)and P(3 and loaded)
P(3 and fair) = P(3 | fair) x P(fair)= (0.166)(0.50) = 0.083
P(3 and loaded) = P(3 | loaded) x P(loaded)
= (0.60)(0.50) = 0.300
-
7/28/2019 PPT Render Ch 2
33/91
2008 Prentice-Hall, Inc. 2 33
Pos terio r Probab i l i t ies
We know the probability of the die being fair orloaded is
P(fair) = 0.50 P(loaded) = 0.50
And that
P(3 | fair) = 0.166 P(3 | loaded) = 0.60
We compute the probabilities ofP(3 and fair)and P(3 and loaded)
P(3 and fair) = P(3 | fair) x P(fair)= (0.166)(0.50) = 0.083
P(3 and loaded) = P(3 | loaded) x P(loaded)
= (0.60)(0.50) = 0.300
The sum of these probabilitiesgives us the unconditionalprobability of tossing a 3
P(3) = 0.083 + 0.300 = 0.383
-
7/28/2019 PPT Render Ch 2
34/91
2008 Prentice-Hall, Inc. 2 34
Pos terio r Probab i l i t ies
P(loaded | 3) = = = 0.78P(loaded and 3)
P(3)
0.300
0.383
The probability that the die was loaded is
P(fair | 3) = = = 0.22P(fair and 3)
P(3)
0.083
0.383
If a 3 does occur, the probability that the die rolledwas the fair one is
These are the revisedorposter iorprobabi l i t iesfor thenext roll of the die
We use these to revise ourpr ior probabi l i tyestimates
-
7/28/2019 PPT Render Ch 2
35/91
2008 Prentice-Hall, Inc. 2 35
Bayes Calcu lat ionsGiven event Bhas occurred
STATE OFNATURE
P(B| STATEOF NATURE)
PRIORPROBABILITY
JOINTPROBABILITY
POSTERIORPROBABILITY
A P(B| A) x P(A) = P(Band A) P(Band A)/P(B) = P(A|B)
A P(B| A) x P(A) = P(Band A) P(Band A)/P(B) = P(A|B)
P(B)
Table 2.2Given a 3 was rolled
STATE OF
NATURE
P(B| STATE
OF NATURE)
PRIOR
PROBABILITY
JOINT
PROBABILITY
POSTERIOR
PROBABILITYFair die 0.166 x 0.5 = 0.083 0.083/0.383 = 0.22
Loaded die 0.600 x 0.5 = 0.300 0.300/0.383 = 0.78
P(3) = 0.383
Table 2.3
-
7/28/2019 PPT Render Ch 2
36/91
2008 Prentice-Hall, Inc. 2 36
General Form of Bayes Theorem
)()|()()|(
)()|(
)|( APABPAPABP
APABP
BAP
We can compute revised probabilities moredirectly by using
where
the complement of the event ;for example, if is the event fair die,
then is loaded die
AA
AA
-
7/28/2019 PPT Render Ch 2
37/91
2008 Prentice-Hall, Inc. 2 37
General Form of Bayes Theorem
This is basically what we did in the previous example
If we replace with fair dieReplace with loaded die
Replace with 3 rolledWe get
AAB
)|( rolled3diefairP
)()|()()|()()|(
loadedloaded3fairfair3fairfair3
PPPPPP
2203830
0830
5006005001660
5001660.
.
.
).)(.().)(.(
).)(.(
-
7/28/2019 PPT Render Ch 2
38/91
2008 Prentice-Hall, Inc. 2 38
FurtherProbab i l ity Revis ions
We can obtain additional information by performingthe experiment a second time
If you can afford it, perform experimentsseveral times
We roll the die again and again get a 3
500loadedand500fair .)(.)( PP
3606060loaded33
027016601660fair33
.).)(.()|,(
.).)(.()|,(
P
P
-
7/28/2019 PPT Render Ch 2
39/91
2008 Prentice-Hall, Inc. 2 39
FurtherProbab i l ity Revis ions
We can obtain additional information by performingthe experiment a second time
If you can afford it, perform experimentsseveral times
We roll the die again and again get a 3
500loadedand500fair .)(.)( PP
3606060loaded33
027016601660fair33
.).)(.()|,(
.).)(.()|,(
P
P
)()|,()( fairfair33fairand3,3 PPP 0130500270 .).)(.(
)()|,()( loadedloaded33loadedand3,3 PPP 18050360 .).)(.(
-
7/28/2019 PPT Render Ch 2
40/91
2008 Prentice-Hall, Inc. 2 40
FurtherProbab i l ity Revis ions
We can obtain additional information by performingthe experiment a second time
If you can afford it, perform experimentsseveral times
We roll the die again and again get a 3
50.0)loaded(and50.0)fair( PP
36.0)6.0)(6.0()loaded|3,3(
027.0)166.0)(166.0()fair|3,3(
P
P
)()|,()( fairfair33fairand3,3 PPP 0130500270 .).)(.(
)()|,()( loadedloaded33loadedand3,3 PPP 18050360 .).)(.(
06701930
0130
33
fairand3,333fair .
.
.
),(
)(),|(
P
PP
93301930
180
33
loadedand3,333loaded .
.
.
),(
)(),|(
P
PP
-
7/28/2019 PPT Render Ch 2
41/91
2008 Prentice-Hall, Inc. 2 41
FurtherProbab i l ity Revis ions
After the first roll of the die
probability the die is fair = 0.22
probability the die is loaded = 0.78
After the second roll of the die
probability the die is fair = 0.067probability the die is loaded = 0.933
-
7/28/2019 PPT Render Ch 2
42/91
2008 Prentice-Hall, Inc. 2 42
Random Variab les
Discrete random variablescan assume onlya finite or limited set of values
Cont inuous random var iablescan assumeany one of an infinite set of values
A random variableassigns a real numberto every possible outcome or event in anexperiment
X= number of refrigerators sold during the day
-
7/28/2019 PPT Render Ch 2
43/91
2008 Prentice-Hall, Inc. 2 43
Random Variab les Numbers
EXPERIMENT OUTCOMERANDOM
VARIABLES
RANGE OFRANDOM
VARIABLES
Stock 50Christmas trees
Number of Christmastrees sold
X 0, 1, 2,, 50
Inspect 600
items
Number of acceptable
items
Y 0, 1, 2,, 600
Send out 5,000sales letters
Number of peopleresponding to theletters
Z 0, 1, 2,, 5,000
Build an
apartmentbuilding
Percent of building
completed after 4months
R 0 R 100
Test the lifetimeof a lightbulb(minutes)
Length of time thebulb lasts up to 80,000minutes
S 0 S 80,000
Table 2.4
-
7/28/2019 PPT Render Ch 2
44/91
2008 Prentice-Hall, Inc. 2 44
Random Variab les Not Numbers
EXPERIMENT OUTCOMERANDOM
VARIABLES
RANGE OFRANDOM
VARIABLES
Studentsrespond to aquestionnaire
Strongly agree (SA)Agree (A)Neutral (N)Disagree (D)Strongly disagree (SD)
5 if SA4 if A..
X= 3 if N..2 if D..1 if SD
1, 2, 3, 4, 5
One machineis inspected
Defective Y=Not defective
0 if defective
1 if not defective
0, 1
Consumersrespond tohow they likea product
GoodAveragePoor
3 if good.Z= 2 if average
1 if poor..
1, 2, 3
Table 2.5
-
7/28/2019 PPT Render Ch 2
45/91
2008 Prentice-Hall, Inc. 2 45
Probabi l i ty Distr ibu t ion o f a
Discrete Random Variable
Dr. Shannon asked studentsto respond to the statement,
The textbook was wellwritten and helped meacquire the necessaryinformation.
Selecting the right probability distributionis important
Fordisc rete random variablesa
probability is assigned to each event
5. Strongly agree4. Agree
3. Neutral2. Disagree1. Strongly disagree
-
7/28/2019 PPT Render Ch 2
46/91
2008 Prentice-Hall, Inc. 2 46
Probabi l i ty Distr ibu t ion o f a
Discrete Random Variable
OUTCOMERANDOMVARIABLE (X)
NUMBERRESPONDING
PROBABILITYP(X)
Strongly agree 5 10 0.1 = 10/100
Agree 4 20 0.2 = 20/100
Neutral 3 30 0.3 = 30/100Disagree 2 30 0.3 = 30/100
Strongly disagree 1 10 0.1 = 10/100
Total 100 1.0 = 100/100
Distribution follows all three rules1. Events are mutually exclusive and collectively exhaustive
2. Individual probability values are between 0 and 1
3. Total of all probability values equals 1
Table 2.6
-
7/28/2019 PPT Render Ch 2
47/91
2008 Prentice-Hall, Inc. 2 47
Probabi l i ty Distr ibut ion fo r
Dr. Shannons Class
P(X)
0.4
0.3
0.2
0.1
0| | | | | |
1 2 3 4 5
XFigure 2.5
-
7/28/2019 PPT Render Ch 2
48/91
2008 Prentice-Hall, Inc. 2 48
Probabi l i ty Distr ibut ion fo r
Dr. Shannons Class
P(X)
0.4
0.3
0.2
0.1
0| | | | | |
1 2 3 4 5
XFigure 2.5
Central tendency of thedistribution is the meanorexpected value
Amount of variability orspread is the variance
-
7/28/2019 PPT Render Ch 2
49/91
2008 Prentice-Hall, Inc. 2 49
Expected Value of a Disc rete
Probabi l i ty Distr ibu t ion
n
iii XPXXE
1 )(...)( 2211 nn XPXXPXXPX
The expected value is a measure of the centraltendencyof the distribution and is a weightedaverage of the values of the random variable
where
iX
)(iXP
n
i 1
)(XE
= random variables possible values
= probability of each of the random variablespossible values
= summation sign indicating we are adding all npossible values
= expected value or mean of the random sample
-
7/28/2019 PPT Render Ch 2
50/91
2008 Prentice-Hall, Inc. 2 50
Variance of a
Discrete Probabi l i ty Distr ibu t ion
For a discrete probability distribution thevariance can be computed by
)()]([ n
iii XPXEX1
22
Variance
where
iX
)(XE
)(i
XP
= random variables possible values
= expected value of the random variable
= difference between each value of the randomvariable and the expected mean
= probability of each possible value of therandom sample
)]([ XEXi
-
7/28/2019 PPT Render Ch 2
51/91
2008 Prentice-Hall, Inc. 2 51
Variance of a
Discrete Probabi l i ty Distr ibu t ion
For Dr. Shannons class
)()]([variance5
1
2
i
iiXPXEX
).().().().(variance 2092410925 22).().().().( 3092230923 22
).().( 10921 2
291
36102430003024204410
.
.....
-
7/28/2019 PPT Render Ch 2
52/91
2008 Prentice-Hall, Inc. 2 52
Variance of a
Discrete Probabi l i ty Distr ibu t ion
A related measure of dispersion is thestandard deviation
2
Variance where
= square root
= standard deviation
-
7/28/2019 PPT Render Ch 2
53/91
2008 Prentice-Hall, Inc. 2 53
Variance of a
Discrete Probabi l i ty Distr ibu t ion
A related measure of dispersion is thestandard deviation
2
Variance where
= square root
= standard deviation
For the textbook question
Variance 141291 ..
-
7/28/2019 PPT Render Ch 2
54/91
2008 Prentice-Hall, Inc. 2 54
Probabi l i ty Distr ibu t ion o f a
Cont inuous Random Var iable
Since random variables can take on an infinitenumber of values, the fundamental rules forcontinuous random variables must be modified
The sum of the probability values must still
equal 1 But the probability of each value of the
random variable must equal 0 or the sumwould be infinitely large
The probability distribution is defined by a
continuous mathematical function called theprobability density function or just the probabilityfunction
Represented by f(X)
-
7/28/2019 PPT Render Ch 2
55/91
2008 Prentice-Hall, Inc. 2 55
Probabi l i ty Distr ibu t ion o f a
Cont inuous Random Var iable
Probability
| | | | | | |
5.06 5.10 5.14 5.18 5.22 5.26 5.30
Weight (grams)
Figure 2.6
-
7/28/2019 PPT Render Ch 2
56/91
2008 Prentice-Hall, Inc. 2 56
The Binom ial Distr ibut ion
Many business experiments can becharacterized by the Bernoulli process
The Bernoulli process is described by the
binomial probability distribution1. Each trial has only two possible outcomes
2. The probability stays the same from one trialto the next
3. The trials are statistically independent4. The number of trials is a positive integer
-
7/28/2019 PPT Render Ch 2
57/91
2008 Prentice-Hall, Inc. 2 57
The Binom ial Distr ibut ion
The binomial distribution is used to find theprobability of a specific number of successesout ofntrials
We need to known= number of trials
p= the probability of success on anysingle trial
We letr= number of successes
q= 1p= the probability of a failure
-
7/28/2019 PPT Render Ch 2
58/91
2008 Prentice-Hall, Inc. 2 58
The Binom ial Distr ibut ion
The binomial formula is
rnrqp
rnr
nnr
)!(!!
trialsinsuccessesofyProbabilit
The symbol ! means factorial, and
n! = n(n 1)(n2)(1)
For example
4! = (4)(3)(2)(1) = 24By definition
1! = 1 and 0! = 1
-
7/28/2019 PPT Render Ch 2
59/91
2008 Prentice-Hall, Inc. 2 59
The Binom ial Distr ibut ion
NUMBER OFHEADS (r) Probability = (0.5)
r(0.5)5r5!
r!(5r)!
0 0.03125 = (0.5)0(0.5)5 0
1 0.15625 = (0.5)1(0.5)5 1
2 0.31250 = (0.5)2(0.5)5 2
3 0.31250 = (0.5)3(0.5)5 3
4 0.15625 = (0.5)4(0.5)5 4
5 0.03125 = (0.5)5(0.5)5 5
5!0!(5 0)!
5!1!(5 1)!
5!2!(5 2)!
5!3!(5 3)!
5!4!(5 4)!
5!5!(5 5)!
Table 2.7
-
7/28/2019 PPT Render Ch 2
60/91
2008 Prentice-Hall, Inc. 2 60
Solv ing Problems w i th the
Binom ial Formula
We want to find the probability of 4 heads in 5 tosses
n= 5, r= 4, p= 0.5, and q= 1 0.5 = 0.5
Thus454 5050
454
5trials5insuccesses4 ..)!(!
!)(P
156250500625011234
12345.).)(.(
)!)()()((
))()()((
Or about 16%
S
-
7/28/2019 PPT Render Ch 2
61/91
2008 Prentice-Hall, Inc. 2 61
Solv ing Problems w i th the
Binom ial Formula
Probability
P(r)
| | | | | | |1 2 3 4 5 6
Values ofr(number of successes)
0.4
0.3
0.2
0.1
0
Figure 2.7
S l i P bl i h
-
7/28/2019 PPT Render Ch 2
62/91
2008 Prentice-Hall, Inc. 2 62
Solv ing Problems w ith
B inom ial TablesMSA Electronics is experimenting with themanufacture of a new transistor
Every hour a sample of 5 transistors is taken
The probability of one transistor being
defective is 0.15What is the probability of finding 3, 4, or 5 defective?
n= 5, p= 0.15, and r= 3, 4, or 5So
We could use the formula to solve this problem,but using the table is easier
S l i P bl i th
-
7/28/2019 PPT Render Ch 2
63/91
2008 Prentice-Hall, Inc. 2 63
Solv ing Problems w ith
B inom ial TablesP
n r 0.05 0.10 0.15
5 0 0.7738 0.5905 0.4437
1 0.2036 0.3281 0.3915
2 0.0214 0.0729 0.1382
3 0.0011 0.0081 0.0244
4 0.0000 0.0005 0.0022
5 0.0000 0.0000 0.0001
Table 2.8 (partial)
We find the three probabilities in the tableforn= 5, p= 0.15, and r= 3, 4, and 5 andadd them together
S l i P bl i th
-
7/28/2019 PPT Render Ch 2
64/91
2008 Prentice-Hall, Inc. 2 64
Table 2.8 (partial)
We find the three probabilities in the tableforn= 5, p= 0.15, and r= 3, 4, and 5 andadd them together
Solv ing Problems w ith
B inom ial TablesP
n r 0.05 0.10 0.15
5 0 0.7738 0.5905 0.4437
1 0.2036 0.3281 0.3915
2 0.0214 0.0729 0.1382
3 0.0011 0.0081 0.0244
4 0.0000 0.0005 0.0022
5 0.0000 0.0000 0.0001
)()()()( 543defectsmoreor3 PPPP 02670000100022002440 ....
S l i P bl i th
-
7/28/2019 PPT Render Ch 2
65/91
2008 Prentice-Hall, Inc. 2 65
Solv ing Problems w ith
B inom ial TablesIt is easy to find the expected value (or mean) andvariance of a binomial distribution
Expected value (mean) = np
Variance = np(1p)
For the MSA example
6375085015051Variance7501505valueExpected
.).)(.()(.).( pnp np
-
7/28/2019 PPT Render Ch 2
66/91
2008 Prentice-Hall, Inc. 2 66
The Normal Distr ibu t ion
The no rmal dis t ribut ionis the most popularand useful continuous probabilitydistribution
The formula for the probability density
function is rather complex
2
2
2
2
1
)(
)(
x
eXf
The normal distribution is specifiedcompletely when we know the mean, ,and the standard deviation,
-
7/28/2019 PPT Render Ch 2
67/91
2008 Prentice-Hall, Inc. 2 67
The Normal Distr ibu t ion
The normal distribution is symmetrical,with the midpoint representing the mean
Shifting the mean does not change the
shape of the distribution Values on the Xaxis are measured in the
number of standard deviations away fromthe mean
As the standard deviation becomes larger,the curve flattens
As the standard deviation becomessmaller, the curve becomes steeper
-
7/28/2019 PPT Render Ch 2
68/91
2008 Prentice-Hall, Inc. 2 68
The Normal Distr ibu t ion
| | |
40 = 50 60
| | |
= 40 50 60
Smaller, same
| | |
40 50 = 60
Larger, same
Figure 2.8
-
7/28/2019 PPT Render Ch 2
69/91
2008 Prentice-Hall, Inc. 2 69
The Normal Distr ibu t ion
Figure 2.9
Same , smaller
Same
, larger
-
7/28/2019 PPT Render Ch 2
70/91
2008 Prentice-Hall, Inc. 2 70
The Normal Distr ibu t ion
Figure 2.10
68%16% 16%
1 +1a b
95.4%2.3% 2.3%
2 +2a b
99.7%0.15% 0.15%
3 +3a b
-
7/28/2019 PPT Render Ch 2
71/91
2008 Prentice-Hall, Inc. 2 71
The Normal Distr ibu t ion
If IQs in the United States were normally distributedwith = 100 and = 15, then
1. 68% of the population would have IQs
between 85 and 115 points (1)2. 95.4% of the people have IQs between 70
and 130 (2)3. 99.7% of the population have IQs in the
range from 55 to 145 points (
3)4. Only 16% of the people have IQs greaterthan 115 points (from the first graph, thearea to the right of +1)
-
7/28/2019 PPT Render Ch 2
72/91
2008 Prentice-Hall, Inc. 2 72
Using the Standard Normal Tab le
Step 1Convert the normal distribution into a standardnormal dis t r ibut ion
A standard normal distribution has a meanof 0 and a standard deviation of 1
The new standard random variable is Z
XZ
whereX= value of the random variable we want to measure
= mean of the distribution
= standard deviation of the distributionZ= number of standard deviations from Xto the mean,
-
7/28/2019 PPT Render Ch 2
73/91
2008 Prentice-Hall, Inc. 2 73
Using the Standard Normal Tab le
For example, = 100, = 15, and we want to findthe probability that Xis less than 130
15
100130 X
Z
devstd215
30
| | | | | | |
55 70 85 100 115 130 145
| | | | | | |
3 2 1 0 1 2 3
X= IQ
XZ
= 100 = 15P(X< 130)
Figure 2.11
-
7/28/2019 PPT Render Ch 2
74/91
2008 Prentice-Hall, Inc. 2 74
Using the Standard Normal Tab le
Step 2Look up the probability from a table of normalcurve areas
Use Appendix A or Table 2.9 (portion below)
The column on the left has Zvalues The row at the top has second decimal
places for the Zvalues
AREA UNDER THE NORMAL CURVE
Z 0.00 0.01 0.02 0.031.8 0.96407 0.96485 0.96562 0.96638
1.9 0.97128 0.97193 0.97257 0.97320
2.0 0.97725 0.97784 0.97831 0.97882
2.1 0.98214 0.98257 0.98300 0.98341
2.2 0.98610 0.98645 0.98679 0.98713
Table 2.9
P(X< 130)= (Z< 2.00)
= 97.7%
-
7/28/2019 PPT Render Ch 2
75/91
2008 Prentice-Hall, Inc. 2 75
Haynes Cons truct ion Company
Haynes builds apartment buildings
Total construction time follows a normaldistribution
For triplexes, = 100 days and = 20 days Contract calls for completion in 125 days Late completion will incur a severe penalty
fee
What is the probability of completing in 125
days?
-
7/28/2019 PPT Render Ch 2
76/91
2008 Prentice-Hall, Inc. 2 76
Haynes Cons truct ion Company
From Appendix A, forZ= 1.25 the area is 0.89435
There is about an 89% probability Hayneswill not violate the contract
20
100125 X
Z
25120
25.
= 100 days X= 125 days
= 20 daysFigure 2.12
-
7/28/2019 PPT Render Ch 2
77/91
2008 Prentice-Hall, Inc. 2 77
Haynes Cons truct ion Company
Haynes builds apartment buildings
Total construction time follows a normaldistribution
For triplexes, = 100 days and = 20 days Completion in 75 days or less will earn a
bonus of $5,000
What is the probability of getting thebonus?
-
7/28/2019 PPT Render Ch 2
78/91
2008 Prentice-Hall, Inc. 2 78
Haynes Cons truct ion Company
But Appendix A has only positive Zvalues, theprobability we are looking for is in the negative tail
20
10075 X
Z
25120
25.
Figure 2.12
= 100 daysX= 75 days
P(X < 75 days)
Area ofInterest
-
7/28/2019 PPT Render Ch 2
79/91
2008 Prentice-Hall, Inc. 2 79
Haynes Cons truct ion Company
Because the curve is symmetrical, we can lookat the probability in the positive tail for thesame distance away from the mean
20
10075 X
Z
25120
25.
= 100 days X= 125 days
P(X > 125 days)Area of
Interest
-
7/28/2019 PPT Render Ch 2
80/91
2008 Prentice-Hall, Inc. 2 80
Haynes Cons truct ion Company
= 100 days X= 125 days
We know the probabilitycompleting in125 days is 0.89435
So the probabilitycompleting in more
than 125 days is1 0.89435 = 0.10565
-
7/28/2019 PPT Render Ch 2
81/91
2008 Prentice-Hall, Inc. 2 81
Haynes Cons truct ion Company
= 100 daysX= 75 days
The probability of completing in less than75 days is 0.10565 or about 11%
Going back tothe left tail of thedistribution
The probabilitycompleting in morethan 125 days is1 0.89435 = 0.10565
-
7/28/2019 PPT Render Ch 2
82/91
2008 Prentice-Hall, Inc. 2 82
Haynes Cons truct ion Company
Haynes builds apartment buildings
Total construction time follows a normaldistribution
For triplexes, = 100 days and = 20 days What is the probability of completing
between 110 and 125 days?
We know the probability of completing in 125
days, P(X< 125) = 0.89435 We have to complete the probability of
completing in 110 days and find the areabetween those two events
C C
-
7/28/2019 PPT Render Ch 2
83/91
2008 Prentice-Hall, Inc. 2 83
Haynes Cons truct ion Company
From Appendix A, forZ= 0.5 the area is 0.69146
P(110 < X< 125) = 0.89435 0.69146 = 0.20289or about 20%
20
100110 X
Z
5020
10.
Figure 2.14
= 100days
125days
= 20 days
110days
-
7/28/2019 PPT Render Ch 2
84/91
2008 Prentice-Hall, Inc. 2 84
TheFDistr ibut ion
A continuous probability distribution
The Fstatistic is the ratio of two sample variances
Fdistributions have two sets of degrees of
freedom Degrees of freedom are based on sample size and
used to calculate the numerator and denominator
df1 = degrees of freedom for the numeratordf2 = degrees of freedom for the denominator
Th F Di ib i
-
7/28/2019 PPT Render Ch 2
85/91
2008 Prentice-Hall, Inc. 2 85
TheFDistr ibut ion
df1 = 5
df2 = 6= 0.05Consider the example:
From Appendix D, we get
F, df1, df2 = F0.05, 5, 6 = 4.39
This means
P(F> 4.39) = 0.05
There is only a 5% probability that Fwill exceed 4.39
Th F Di t ib t i
-
7/28/2019 PPT Render Ch 2
86/91
2008 Prentice-Hall, Inc. 2 86
F
TheFDistr ibut ion
Figure 2.15
Th F Di t ib t i
-
7/28/2019 PPT Render Ch 2
87/91
2008 Prentice-Hall, Inc. 2 87
TheFDistr ibut ion
Figure 2.16
F= 4.390.05
Fvalue for 0.05 probabilitywith 5 and 6 degrees offreedom
Th E ti l Di t ib t i
-
7/28/2019 PPT Render Ch 2
88/91
2008 Prentice-Hall, Inc. 2 88
The Exponen t ial Distr ibu t ion
The exponent ial dist r ibu t ion(also calledthe negat ive exponent ial d ist r ibut ion) is acontinuous distribution often used inqueuing models to describe the time
required to service a customerx
eXf )(
where
X= random variable (service times)= average number of units the service facility can
handle in a specific period of time
e= 2.718 (the base of natural logarithms)
Th E ti l Di t ib t i
-
7/28/2019 PPT Render Ch 2
89/91
2008 Prentice-Hall, Inc. 2 89
The Exponen t ial Distr ibu t ion
timeserviceAverage1valueExpected 2
1Variance
f(X)
XFigure 2.17
Th P i Di t ib t i
-
7/28/2019 PPT Render Ch 2
90/91
2008 Prentice-Hall, Inc. 2 90
The Poisson Distr ibut ion
The Poissondis t r ibut ionis a discretedistribution that is often used in queuingmodels to describe arrival rates over time
!)(
XeXP
x
where
P(X) = probability of exactly X arrivals or occurrences
= average number of arrivals per unit of time(the mean arrival rate)
e= 2.718, the base of natural logarithms
X= specific value (0, 1, 2, 3, and so on) of the randomvariable
Th P i Di t ib t i
-
7/28/2019 PPT Render Ch 2
91/91
The Poisson Distr ibut ion
The mean and variance of the distribution areboth
Expected value = Variance =
0.3
0.2
0.1
0| | | | | | | | | |
P(X)