chapter 18 introduction to decision analysis
DESCRIPTION
Business Statistics: A Decision-Making Approach 6 th Edition. Chapter 18 Introduction to Decision Analysis. Chapter Goals. After completing this chapter, you should be able to: Describe the decision environments of certainty and uncertainty - PowerPoint PPT PresentationTRANSCRIPT
Chap 18-1
Business Statistics: A Decision-Making Approach
6th Edition
Chapter 18Introduction to Decision
Analysis
Chap 17-2
Chapter Goals
After completing this chapter, you should be able to:
Describe the decision environments of certainty and uncertainty
Construct a payoff table and an opportunity-loss table
Define and apply the expected value criterion for decision making
Compute the value of perfect information
Develop and use decision trees for decision making
Chap 17-3
Decision Making Overview
Decision Making
Certainty Nonprobabilistic
Uncertainty Probabilistic
Decision Environment Decision Criteria
Chap 17-4
The Decision Environment
Certainty
Uncertainty
Decision Environment Certainty: The results of decision alternatives are known
Example:
Must print 10,000 color brochures
Offset press A: $2,000 fixed cost + $.24 per page
Offset press B: $3,000 fixed cost + $.12 per page
*
Chap 17-5
The Decision Environment
Uncertainty
Certainty
Decision EnvironmentUncertainty: The outcome that will occur after a choice is unknown
Example:
You must decide to buy an item now or wait. If you buy now the price is $2,000. If you wait the price may drop to $1,500 or rise to $2,200. There also may be a new model available later with better features.
*
(continued)
Chap 17-6
Decision Criteria
Nonprobabilistic
Probabilistic
Decision CriteriaNonprobabilistic Decision Criteria: Decision rules that can be applied if the probabilities of uncertain events are not known. *
maximax criterion
maximin criterion
minimax regret criterion
Chap 17-7
Nonprobabilistic
Probabilistic
Decision Criteria
*
Probabilistic Decision Criteria: Consider the probabilities of uncertain events and select an alternative to maximize the expected payoff of minimize the expected loss
maximize expected value
minimize expected opportunity loss
Decision Criteria(continued)
Chap 17-8
A Payoff Table
A payoff table shows alternatives, states of nature, and payoffs
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factory
Average factory
Small factory
200
90
40
50
120
30
-120
-30
20
Chap 17-9
Maximax Solution
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
1.
Maximum Profit
200120 40
The maximax criterion (an optimistic approach): 1. For each option, find the maximum payoff
Chap 17-10
Maximax Solution
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
1.
Maximum Profit
200120 40
The maximax criterion (an optimistic approach): 1. For each option, find the maximum payoff
2. Choose the option with the greatest maximum payoff
2.
Greatest maximum
is to choose Large
factory
(continued)
Chap 17-11
R program for Maximax act1=c(200,50,-120) #data for action 1 act2=c(90,120,-30) act3=c(40,30,20) nact=3 #number of actions nj=3 #number of states of nature payoff=matrix(rbind(act1,act2,act3,act4),nact,nj) payoff maxact=rep(0,nact) #initialize i=1 while (i<=nact) { maxact[i]=max(payoff[i,]) i=i+1} maxact maximax=max(maxact) #optimistic criterion maximax maximax
Chap 17-12
Maximin Solution
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
1.
Minimum Profit
-120 -30 20
The maximin criterion (a pessimistic approach): 1. For each option, find the minimum payoff
Chap 17-13
Maximin Solution
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
1.
Minimum Profit
-120 -30 20
The maximin criterion (a pessimistic approach): 1. For each option, find the minimum payoff
2. Choose the option with the greatest minimum payoff
2.
Greatest minimum
is to choose Small
factory
(continued)
Chap 17-14
R for Maximin (payoff in memory)
minact=rep(0,nact) #initialize i=1 while (i<=nact) { minact[i]=min(payoff[i,]) i=i+1} minact #minimum payoffs for each action maximin=max(minact) #pessimistic criterion maximin maximin
Chap 17-15
Opportunity Loss
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20The choice “Average factory” has payoff 90 for “Strong Economy”. Given
“Strong Economy”, the choice of “Large factory” would have given a payoff of 200, or 110 higher. Opportunity loss = 110 for this cell.
Opportunity loss is the difference between an actual payoff for a decision and the optimal payoff for that state of nature (in each column)
Payoff Table
Chap 17-16
Opportunity Loss
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
(continued)
Investment Choice
(Alternatives)
Opportunity Loss in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
0110160
70 0
90
140500
Payoff Table
Opportunity Loss Table
Chap 17-17
R for opportunity loss matrix
opp=payoff #initialize old info in memory of R j=1 while (j<=nj) { opp[1:nact,j]= max(payoff[,j]) j=j+1} opp #the opportunity table opploss=opp-payoff #opportunity loss table opploss
Chap 17-18
Minimax Regret Solution
Investment Choice
(Alternatives)
Opportunity Loss in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
0110160
70 0
90
140500
Opportunity Loss Table
The minimax regret criterion:1. For each alternative, find the maximum opportunity
loss (or “regret”)
1.
Maximum Op. Loss
140110160
Chap 17-19
R for maximum regret
#assume previous info is in memory of R maxreg=rep(0,nact) #initialize i=1 while (i<=nact) { maxreg[i]= max(opploss[i,]) i=i+1} maxreg #maximum regrets
Chap 17-20
Minimax Regret Solution
Investment Choice
(Alternatives)
Opportunity Loss in $1,000’s
(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Large factoryAverage factorySmall factory
0110160
70 0
90
140500
Opportunity Loss Table
The minimax regret criterion:1. For each alternative, find the maximum opportunity
loss (or “regret”)
2. Choose the option with the smallest maximum loss
1.
Maximum Op. Loss
140110160
2.
Smallest maximum loss is to choose
Average factory
(continued)
Chap 17-21
Expected Value Solution
The expected value is the weighted average payoff, given specified probabilities for each state of nature
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
(.3)
Stable Economy
(.5)
Weak Economy
(.2)
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
Suppose these probabilities have been assessed for these states of nature
Chap 17-22
Minimax regret+Expected Value
minimaxregret=min(maxreg) minimaxregret prob=c(0.3,0.5,0.2) length(prob) # Error if number of probabilities are not equal to number of states
of nature nj sum(prob) # sum probabilities should be one evalu=payoff %*%prob evalu maxevalu=max(evalu) maxevalu
Chap 17-23
Expected Value Solution
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
(.3)
Stable Economy
(.5)
Weak Economy
(.2)
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
Example: EV (Average factory) = 90(.3) + 120(.5) + (-30)(.2)
= 81
Expected Values
618131
Maximize expected value by choosing Average factory
(continued)
Chap 17-24
Expected Opportunity Loss Solution
Investment Choice
(Alternatives)
Opportunity Loss in $1,000’s
(States of Nature)
Strong Economy
(.3)
Stable Economy
(.5)
Weak Economy
(.2)
Large factoryAverage factorySmall factory
0110160
70 0
90
140500
Example: EOL (Large factory) = 0(.3) + 70(.5) + (140)(.2)
= 63
Expected Op. Loss
(EOL)634393
Minimize expected
op. loss by choosing Average factory
Opportunity Loss Table
Chap 17-25
Expected Opportunity Loss in R
expopploss=opploss%*%prob expopploss minexpopploss=min(expopploss) Minexpopploss #min of exp opp loss
Chap 17-26
Cost of Uncertainty
Cost of Uncertainty (also called Expected Value of Perfect Information, or EVPI)
Cost of Uncertainty
= Expected Value Under Certainty (EVUC)
– Expected Value without information (EV)
so: EVPI = EVUC – EV
Chap 17-27
Expected Value Under Certainty
Expected Value Under Certainty (EVUC):
EVUC = expected value of the best decision, given perfect information
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
(.3)
Stable Economy
(.5)
Weak Economy
(.2)
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
Example: Best decision given “Strong Economy” is “Large factory”
200 120 20
Chap 17-28
Expected Value Under Certainty
Investment Choice
(Alternatives)
Profit in $1,000’s
(States of Nature)
Strong Economy
(.3)
Stable Economy
(.5)
Weak Economy
(.2)
Large factoryAverage factorySmall factory
2009040
50120 30
-120-30 20
200 120 20
(continued)
EVUC = 200(.3)+120(.5)+20(.2) = 124
Now weight these outcomes with their probabilities to find EVUC:
Chap 17-29
Cost of Uncertainty Solution
Cost of Uncertainty (EVPI)
= Expected Value Under Certainty (EVUC)
– Expected Value without information (EV)
so: EVPI = EVUC – EV = 124 – 81 = 43
Recall: EVUC = 124
EV is maximized by choosing “Average factory”, where EV = 81
Chap 17-30
Decision Tree Analysis
A Decision tree shows a decision problem, beginning with the initial decision and ending will all possible outcomes and payoffs.
Use a square to denote decision nodes
Use a circle to denote uncertain events
Chap 17-31
Sample Decision Tree
Large factory
Small factory
Average factory
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
Chap 17-32
Add Probabilities and Payoffs
Large factory
Small factory
Decision
Average factory
Uncertain Events(States of Nature)
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
(continued)
PayoffsProbabilities
200
50
-120
40
30
20
90
120
-30
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
Chap 17-33
Fold Back the Tree
Large factory
Small factory
Average factory
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
200
50
-120
40
30
20
90
120
-30
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
EV=200(.3)+50(.5)+(-120)(.2)=61
EV=90(.3)+120(.5)+(-30)(.2)=81
EV=40(.3)+30(.5)+20(.2)=31
Chap 17-34
Make the Decision
Large factory
Small factory
Average factory
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
Strong Economy
Stable Economy
Weak Economy
200
50
-120
40
30
20
90
120
-30
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
(.3)
(.5)
(.2)
EV=61
EV=81
EV=31
Maximum
EV=81
Chap 17-35
Chapter Summary
Examined decision making environments certainty and uncertainty
Reviewed decision making criteria nonprobabilistic: maximax, maximin, minimax regret probabilistic: expected value, expected opp. loss
Computed the Cost of Uncertainty (EVPI) Developed decision trees and applied them to
decision problems