decision analysis
DESCRIPTION
SPPT Course SampleTRANSCRIPT
1 1 Slide
Slide
© 2005 Thomson/South-Western
Chapter 13Decision Analysis
Problem Formulation Decision Making without
Probabilities Decision Making with
Probabilities Risk Analysis and Sensitivity
Analysis Decision Analysis with Sample
Information Computing Branch Probabilities
2 2 Slide
Slide
© 2005 Thomson/South-Western
Problem Formulation
A decision problem is characterized by decision alternatives, states of nature, and resulting payoffs.
The decision alternatives are the different possible strategies the decision maker can employ.
The states of nature refer to future events, not under the control of the decision maker, which may occur. States of nature should be defined so that they are mutually exclusive and collectively exhaustive.
3 3 Slide
Slide
© 2005 Thomson/South-Western
Influence Diagrams
An influence diagram is a graphical device showing the relationships among the decisions, the chance events, and the consequences.
Squares or rectangles depict decision nodes.
Circles or ovals depict chance nodes. Diamonds depict consequence nodes. Lines or arcs connecting the nodes show
the direction of influence.
4 4 Slide
Slide
© 2005 Thomson/South-Western
Payoff Tables
The consequence resulting from a specific combination of a decision alternative and a state of nature is a payoff.
A table showing payoffs for all combinations of decision alternatives and states of nature is a payoff table.
Payoffs can be expressed in terms of profit, cost, time, distance or any other appropriate measure.
5 5 Slide
Slide
© 2005 Thomson/South-Western
Decision Trees
A decision tree is a chronological representation of the decision problem.
Each decision tree has two types of nodes; round nodes correspond to the states of nature while square nodes correspond to the decision alternatives.
6 6 Slide
Slide
© 2005 Thomson/South-Western
The branches leaving each round node represent the different states of nature while the branches leaving each square node represent the different decision alternatives.
At the end of each limb of a tree are the payoffs attained from the series of branches making up that limb.
7 7 Slide
Slide
© 2005 Thomson/South-Western
Decision Making without Probabilities
Three commonly used criteria for decision making when probability information regarding the likelihood of the states of nature is unavailable are: • the optimistic approach• the conservative approach• the minimax regret approach.
8 8 Slide
Slide
© 2005 Thomson/South-Western
Optimistic Approach
The optimistic approach would be used by an optimistic decision maker.
The decision with the largest possible payoff is chosen.
If the payoff table was in terms of costs, the decision with the lowest cost would be chosen.
9 9 Slide
Slide
© 2005 Thomson/South-Western
Conservative Approach
The conservative approach would be used by a conservative decision maker.
For each decision the minimum payoff is listed and then the decision corresponding to the maximum of these minimum payoffs is selected. (Hence, the minimum possible payoff is maximized.)
If the payoff was in terms of costs, the maximum costs would be determined for each decision and then the decision corresponding to the minimum of these maximum costs is selected. (Hence, the maximum possible cost is minimized.)
10 10 Slide
Slide
© 2005 Thomson/South-Western
Minimax Regret Approach
The minimax regret approach requires the construction of a regret table or an opportunity loss table.
This is done by calculating for each state of nature the difference between each payoff and the largest payoff for that state of nature.
Then, using this regret table, the maximum regret for each possible decision is listed.
The decision chosen is the one corresponding to the minimum of the maximum regrets.
11 11 Slide
Slide
© 2005 Thomson/South-Western
Example
Consider the following problem with three decision alternatives and three states of nature with the following payoff table representing profits:
States of Nature s1 s2 s3
d1 4 4 -2
Decisions d2 0 3 -1
d3 1 5 -3
12 12 Slide
Slide
© 2005 Thomson/South-Western
Example: Optimistic Approach
An optimistic decision maker would use the optimistic (maximax) approach. We choose the decision that has the largest single value in the payoff table.
Maximum Decision Payoff
d1 4
d2 3
d3 5
Maximaxpayoff
Maximaxdecision
13 13 Slide
Slide
© 2005 Thomson/South-Western
Example: Optimistic Approach
Solution SpreadsheetA B C D E F
123 Decision Maximum Recommended4 Alternative s1 s2 s3 Payoff Decision5 d1 4 4 -2 46 d2 0 3 -1 37 d3 1 5 -3 5 d389 5
State of Nature
Best Payoff
PAYOFF TABLE
14 14 Slide
Slide
© 2005 Thomson/South-Western
Example: Conservative Approach
A conservative decision maker would use the conservative (maximin) approach. List the minimum payoff for each decision. Choose the decision with the maximum of these minimum payoffs.
Minimum Decision Payoff
d1 -2
d2 -1
d3 -3
Maximindecision
Maximinpayoff
15 15 Slide
Slide
© 2005 Thomson/South-Western
Example: Conservative Approach
Solution SpreadsheetA B C D E F
123 Decision Minimum Recommended4 Alternative s1 s2 s3 Payoff Decision5 d1 4 4 -2 -26 d2 0 3 -1 -1 d27 d3 1 5 -3 -389 -1
State of Nature
Best Payoff
PAYOFF TABLE
16 16 Slide
Slide
© 2005 Thomson/South-Western
For the minimax regret approach, first compute a regret table by subtracting each payoff in a column from the largest payoff in that column. In this example, in the first column subtract 4, 0, and 1 from 4; etc. The resulting regret table is:
s1 s2 s3
d1 0 1 1
d2 4 2 0
d3 3 0 2
Example: Minimax Regret Approach
17 17 Slide
Slide
© 2005 Thomson/South-Western
For each decision list the maximum regret. Choose the decision with the minimum of these values.
Maximum Decision Regret d1 1
d2 4
d3 3
Example: Minimax Regret Approach
Minimaxdecision
Minimaxregret
18 18 Slide
Slide
© 2005 Thomson/South-Western
Solution SpreadsheetA B C D E F
12 Decision 3 Alternative s1 s2 s34 d1 4 4 -25 d2 0 3 -16 d3 1 5 -3789 Decision Maximum Recommended10 Alternative s1 s2 s3 Regret Decision11 d1 0 1 1 1 d112 d2 4 2 0 413 d3 3 0 2 314 1Minimax Regret Value
State of NaturePAYOFF TABLE
State of NatureOPPORTUNITY LOSS TABLE
Example: Minimax Regret Approach
19 19 Slide
Slide
© 2005 Thomson/South-Western
Decision Making with Probabilities
Expected Value Approach• If probabilistic information regarding the
states of nature is available, one may use the expected value (EV) approach.
• Here the expected return for each decision is calculated by summing the products of the payoff under each state of nature and the probability of the respective state of nature occurring.
• The decision yielding the best expected return is chosen.
20 20 Slide
Slide
© 2005 Thomson/South-Western
The expected value of a decision alternative is the sum of weighted payoffs for the decision alternative.
The expected value (EV) of decision alternative di is defined as:
where: N = the number of states of nature P(sj ) = the probability of state of
nature sj
Vij = the payoff corresponding to decision alternative di and state of nature sj
Expected Value of a Decision Alternative
EV( ) ( )d P s Vi j ijj
N
1
EV( ) ( )d P s Vi j ijj
N
1
21 21 Slide
Slide
© 2005 Thomson/South-Western
Example: Burger Prince
Burger Prince Restaurant is considering opening a new restaurant on Main Street. It has three
different models, each with a different
seating capacity. Burger Prince
estimates that the average number of
customers per hour will be 80, 100, or
120. The payoff table for the three
models is on the next slide.
22 22 Slide
Slide
© 2005 Thomson/South-Western
Payoff Table
Average Number of Customers Per Hour
s1 = 80 s2 = 100 s3 = 120
Model A $10,000 $15,000 $14,000
Model B $ 8,000 $18,000 $12,000
Model C $ 6,000 $16,000 $21,000
23 23 Slide
Slide
© 2005 Thomson/South-Western
Expected Value Approach
Calculate the expected value for each decision. The decision tree on the next slide can assist in this calculation. Here d1, d2, d3
represent the decision alternatives of models A, B, C, and s1, s2, s3 represent the states of
nature of 80, 100, and 120.
24 24 Slide
Slide
© 2005 Thomson/South-Western
Decision Tree
11
.2
.4
.4
.4
.2
.4
.4
.2
.4
d1
d2
d3
s1
s1
s1
s2
s3
s2
s2
s3
s3
Payoffs10,000
15,000
14,0008,000
18,000
12,000
6,000
16,000
21,000
22
33
44
25 25 Slide
Slide
© 2005 Thomson/South-Western
Expected Value for Each Decision
Choose the model with largest EV, Model C.
33
d1
d2
d3
EMV = .4(10,000) + .2(15,000) + .4(14,000) = $12,600
EMV = .4(8,000) + .2(18,000) + .4(12,000) = $11,600
EMV = .4(6,000) + .2(16,000) + .4(21,000) = $14,000
Model A
Model B
Model C
22
11
44
26 26 Slide
Slide
© 2005 Thomson/South-Western
Solution SpreadsheetA B C D E F
123 Decision Expected Recommended4 Alternative s1 = 80 s2 = 100 s3 = 120 Value Decision5 d1 = Model A 10,000 15,000 14,000 126006 d2 = Model B 8,000 18,000 12,000 116007 d3 = Model C 6,000 16,000 21,000 14000 d3 = Model C8 Probability 0.4 0.2 0.49 14000
State of Nature
Maximum Expected Value
PAYOFF TABLE
Expected Value Approach
27 27 Slide
Slide
© 2005 Thomson/South-Western
Expected Value of Perfect Information
Frequently information is available which can improve the probability estimates for the states of nature.
The expected value of perfect information (EVPI) is the increase in the expected profit that would result if one knew with certainty which state of nature would occur.
The EVPI provides an upper bound on the expected value of any sample or survey information.
28 28 Slide
Slide
© 2005 Thomson/South-Western
Expected Value of Perfect Information
EVPI Calculation• Step 1:
Determine the optimal return corresponding to each state of nature.
• Step 2: Compute the expected value of these
optimal returns.• Step 3:
Subtract the EV of the optimal decision from the amount determined in step (2).
29 29 Slide
Slide
© 2005 Thomson/South-Western
Calculate the expected value for the optimum payoff for each state of nature and subtract the EV of the optimal decision.
EVPI= .4(10,000) + .2(18,000) + .4(21,000) - 14,000 = $2,000
Expected Value of Perfect Information
30 30 Slide
Slide
© 2005 Thomson/South-Western
SpreadsheetA B C D E F
123 Decision Expected Recommended4 Alternative s1 = 80 s2 = 100 s3 = 120 Value Decision5 d1 = Model A 10,000 15,000 14,000 126006 d2 = Model B 8,000 18,000 12,000 116007 d3 = Model C 6,000 16,000 21,000 14000 d3 = Model C8 Probability 0.4 0.2 0.49 140001011 EVwPI EVPI12 10,000 18,000 21,000 16000 2000
State of Nature
Maximum Expected Value
PAYOFF TABLE
Maximum Payoff
Expected Value of Perfect Information
31 31 Slide
Slide
© 2005 Thomson/South-Western
Risk Analysis
Risk analysis helps the decision maker recognize the difference between:• the expected value of a decision alternative,
and• the payoff that might actually occur
The risk profile for a decision alternative shows the possible payoffs for the decision alternative along with their associated probabilities.
32 32 Slide
Slide
© 2005 Thomson/South-Western
Risk Profile
Model C Decision Alternative
.10.10
.20.20
.30.30
.40.40
.50.50
5 10 15 20 255 10 15 20 25
Pro
bab
ility
Pro
bab
ility
Profit ($thousands)Profit ($thousands)
33 33 Slide
Slide
© 2005 Thomson/South-Western
Sensitivity Analysis
Sensitivity analysis can be used to determine how changes to the following inputs affect the recommended decision alternative:• probabilities for the states of nature• values of the payoffs
If a small change in the value of one of the inputs causes a change in the recommended decision alternative, extra effort and care should be taken in estimating the input value.
34 34 Slide
Slide
© 2005 Thomson/South-Western
Bayes’ Theorem and Posterior Probabilities
Knowledge of sample (survey) information can be used to revise the probability estimates for the states of nature.
Prior to obtaining this information, the probability estimates for the states of nature are called prior probabilities.
With knowledge of conditional probabilities for the outcomes or indicators of the sample or survey information, these prior probabilities can be revised by employing Bayes' Theorem.
The outcomes of this analysis are called posterior probabilities or branch probabilities for decision trees.
35 35 Slide
Slide
© 2005 Thomson/South-Western
Computing Branch Probabilities
Branch (Posterior) Probabilities Calculation • Step 1:
For each state of nature, multiply the prior probability by its conditional probability for the indicator -- this gives the joint probabilities for the states and indicator.
36 36 Slide
Slide
© 2005 Thomson/South-Western
Computing Branch Probabilities
Branch (Posterior) Probabilities Calculation • Step 2:
Sum these joint probabilities over all states -- this gives the marginal probability for the indicator.
• Step 3: For each state, divide its joint
probability by the marginal probability for the indicator -- this gives the posterior probability distribution.
37 37 Slide
Slide
© 2005 Thomson/South-Western
Expected Value of Sample Information
The expected value of sample information (EVSI) is the additional expected profit possible through knowledge of the sample or survey information.
38 38 Slide
Slide
© 2005 Thomson/South-Western
Expected Value of Sample Information
EVSI Calculation• Step 1:
Determine the optimal decision and its expected return for the possible outcomes of the sample using the posterior probabilities for the states of nature.
• Step 2: Compute the expected value of these optimal returns.
• Step 3: Subtract the EV of the optimal
decision obtained without using the sample information from the amount determined in step (2).
39 39 Slide
Slide
© 2005 Thomson/South-Western
Efficiency of Sample Information
Efficiency of sample information is the ratio of EVSI to EVPI.
As the EVPI provides an upper bound for the EVSI, efficiency is always a number between 0 and 1.
40 40 Slide
Slide
© 2005 Thomson/South-Western
Burger Prince must decide whether or not to purchase a marketing survey from Stanton Marketing for $1,000. The results of the survey are "favorable" or "unfavorable". The conditional probabilities are:
P(favorable | 80 customers per hour) = .2
P(favorable | 100 customers per hour) = .5
P(favorable | 120 customers per hour) = .9
Should Burger Prince have the survey performed by Stanton Marketing?
Sample Information
41 41 Slide
Slide
© 2005 Thomson/South-Western
Influence Diagram
RestaurantSize Profit
Avg. Numberof Customers
Per Hour
MarketSurveyResults
MarketSurvey
DecisionChanceConsequence
42 42 Slide
Slide
© 2005 Thomson/South-Western
Favorable
State Prior Conditional Joint Posterior
80 .4 .2 .08 .148
100 .2 .5 .10 .185
120 .4 .9 .36 .667 Total .54 1.000
P(favorable) = .54
Posterior Probabilities
43 43 Slide
Slide
© 2005 Thomson/South-Western
Unfavorable
State Prior Conditional Joint Posterior
80 .4 .8 .32 .696
100 .2 .5 .10 .217
120 .4 .1 .04 .087
Total .46 1.000
P(unfavorable) = .46
Posterior Probabilities
44 44 Slide
Slide
© 2005 Thomson/South-Western
Solution SpreadsheetA B C D E
12 Prior Conditional Joint Posterior3 State of Nature Probabilities Probabilities Probabilities Probabilities4 s1 = 80 0.4 0.2 0.08 0.1485 s2 = 100 0.2 0.5 0.10 0.1856 s3 = 120 0.4 0.9 0.36 0.6677 0.548910 Prior Conditional Joint Posterior11 State of Nature Probabilities Probabilities Probabilities Probabilities12 s1 = 80 0.4 0.8 0.32 0.69613 s2 = 100 0.2 0.5 0.10 0.21714 s3 = 120 0.4 0.1 0.04 0.08715 0.46
Market Research Favorable
P(Favorable) =
Market Research Unfavorable
P(Favorable) =
Posterior Probabilities
45 45 Slide
Slide
© 2005 Thomson/South-Western
Decision Tree
Top Half
s1 (.148)
s1 (.148)
s1 (.148)s2 (.185)
s2 (.185)
s2 (.185)
s3 (.667)
s3 (.667)
s3 (.667)
$10,000
$15,000$14,000$8,000
$18,000
$12,000$6,000
$16,000
$21,000
I1(.54)
d1
d2
d3
22
44
55
66
11
46 46 Slide
Slide
© 2005 Thomson/South-Western
Bottom Half
s1 (.696)
s1 (.696)
s1 (.696)
s2 (.217)
s2 (.217)
s2 (.217)
s3 (.087)
s3 (.087)
s3 (.087)
$10,000
$15,000
$18,000
$14,000$8,000
$12,000$6,000
$16,000
$21,000
I2(.46) d1
d2
d3
77
99
8833
11
Decision Tree
47 47 Slide
Slide
© 2005 Thomson/South-Western
I2(.46)
d1
d2
d3
EMV = .696(10,000) + .217(15,000) +.087(14,000)= $11,433
EMV = .696(8,000) + .217(18,000) + .087(12,000) = $10,554
EMV = .696(6,000) + .217(16,000) +.087(21,000) = $9,475
I1(.54)
d1
d2
d3
EMV = .148(10,000) + .185(15,000) + .667(14,000) = $13,593
EMV = .148 (8,000) + .185(18,000) + .667(12,000) = $12,518
EMV = .148(6,000) + .185(16,000) +.667(21,000) = $17,855
44
55
66
77
88
99
22
33
11
$17,855
$11,433
Decision Tree
48 48 Slide
Slide
© 2005 Thomson/South-Western
If the outcome of the survey is "favorable”,choose Model C. If it is “unfavorable”, choose Model A.
EVSI = .54($17,855) + .46($11,433) - $14,000 = $900.88
Since this is less than the cost of the survey, the survey should not be purchased.
Expected Value of Sample Information
49 49 Slide
Slide
© 2005 Thomson/South-Western
Efficiency of Sample Information
The efficiency of the survey:
EVSI/EVPI = ($900.88)/($2000) = .4504
50 50 Slide
Slide
© 2005 Thomson/South-Western
Bayes’ Decision Rule:
Using the best available estimates of the
probabilities of the respective states of nature
(currently the prior probabilities), calculate the
expected value of the payoff for each of the
possible actions. Choose the action with the
maximum expected payoff.
51 51 Slide
Slide
© 2005 Thomson/South-Western
Bayes’ theorySi: State of Nature (i = 1 ~ n)
P(Si): Prior Probability
Ij: Professional Information (Experiment)( j = 1 ~ n)
P(Ij | Si): Conditional Probability
P(Ij Si) = P(Si Ij): Joint Probability
P(Si | Ij): Posterior Probability
P(Si | Ij)
n
1iiij
iij
j
ji
)S(P)S|I(P
)S(P)S|I(P
)I(P
)IS(P
52 52 Slide
Slide
© 2005 Thomson/South-Western
Home Work
Problem 13-10
Problem 13-21
Due Date: Nov 11, 2008