Decision Analysis Ba 182

Chapter 9
DECISION ANALYSIS
CHAPTER 9
DECISION ANALYSIS
Review Questions
9.1-1 The decision alternatives are to drill for oil or to sell the land.
9.1-2 The consulting geologist believes that there is 1 chance in 4 of oil on the tract of land.
9.1-3 Max does not put much faith in the assessment.
9.1-4 A detailed seismic survey of the land could be done to obtain more information.
9.1-5 The possible states of nature are the possible outcomes of the random factors that affect the
payoff that would be obtained from a decision alternative.
9.1-6 Prior probabilities are the estimated probabilities of the states of nature prior to obtaining
additional information through a test or survey.
9.1-7 The payoffs are quantitative measures of the outcomes from a decision alternative and a
state of nature. Payoffs are generally expressed in monetary terms.
9.2-1 The maximax criterion identifies the maximum payoff for each decision alternative and
chooses the decision alternative with the maximum of these maximum payoffs. The
maximax criterion is for the eternal optimist.
9.2-2 The maximax citerion completely ignores the prior probabilities and ignores all payoffs
except for the largest one.
9.2-3 The maximin criterion identifies the minimum payoff for each decision alternative and
chooses the decision alternative with the maximum of these minimum payoffs. The
maximin criterion is for the total pessimist.
9.2-4 The maximin criterion ignores the prior probabilities and ignores all payoffs except the
maximin payoff.
9.2-5 The maximum likelihood criterion focuses on the most likely state of nature, the one with
the largest prior probability.
9.2-6 Criticisms of the maximum likelihood criterion include: 1) this criterion chooses an
alternative without considering its payoffs for states of nature other than the most likely
one, 2) for alternatives that are not chosen, this criterion ignores their payoffs for states of
nature other than the most likely one, 3) if the differences in the payoffs for the most likely
state of nature are much less than for another somewhat likely state of nature, then it might
make sense to focus on this latter state of nature instead, and 4) if there are many states of
nature and they are nearly equally likely, then the probability that the most likely state of
nature will be the true one is fairly low.
9.2-7 Bayes’ decision rule says to choose the alternative with the largest expected payoff.
Chapter 9 - Page 1
Chapter 9
DECISION ANALYSIS
9.2-8 The expected payoff is calculated by multiplying each payoff by the prior probability of
the corresponding state of nature and then summing these products.
9.2-9 Criticisms of Bayes’ decision rule include: 1) there usually is considerable uncertainty
involved in assigning values to prior probabilities, 2) prior probabilities inherently are at
least largely subjective in nature, whereas sound decision making should be based on
objective data and procedures, and 3) by focusing on average outcomes, expected payoffs
ignore the effect that the amount of variability in the possible outcomes should have on the
decision making.
9.3-1 A decision tree is a graphical display of the progression of decisions and random events to
be considered.
9.3-2 A decision node indicates that a decision needs to be made at that point in the process. An
event node indicates that a random event occurs at that point.
9.3-3 Decision nodes are represented by squares while circles represent event nodes.
9.4-1 Sensitivity analysis might be helpful to study the effect if some of the numbers included in
the model are not correct.
9.4-2 It assures that each piece of data is in only one place and it makes it easy for anyone to
interpret the model, even if they don’t understand TreePlan or decision trees.
9.4-3 A data table displays results of selected output cells for various trial values of a data cell.
9.4-4 If there is less than a 23.75% chance of oil, they should sell. If it’s more, they should drill.
9.5-1 Perfect information means knowing for sure which state of nature is the true state of
nature.
9.5-2 The expected payoff with perfect information is calculated by multiplying the maximum
payoff for each alternative by the prior probability of the corresponding state of nature.
9.5-3 The decision tree should be started with a chance node whose branches are the various
states of nature.
9.5-4 EVPI = EP (with perfect information) – EP (without more information)
9.5-5 If the cost of obtaining more information is more than the expected value of perfect
information then it is not worthwhile to obtain more information.
9.5-6 If the cost of obtaining more information is less than the expected value of perfect
information then it might be worthwhile to obtain more information.
9.5-7 In the Goferbroke problem the EVPI >C so it might be worthwhile to do the seismic
survey.
9.6-1 Posterior probabilities are revised probabilities of the states of nature after doing a test or
survey to improve the prior probabilities.
9.6-2 The possible findings are favorable with oil being fairly likely, or unfavorable with oil
being quite unlikely.
9.6-3 Conditional probabilities need to be estimated.
Chapter 9 - Page 2
Chapter 9
DECISION ANALYSIS
9.6-4 The five kinds of probabilities considered are prior, conditional, joint, unconditional, and
posterior.
9.6-5 P(state and finding) = P(state) * P(finding | state).
9.6-6 P(finding) = sum of P(state and finding) for each state.
9.6-7 P(state | finding) = P(state and finding) / P(finding).
9.6-8 Bayes’ theorem is used to calculate posterior probabilities.
9.7-1 A decision tree provides a graphical display of the progression of decisions and random
events for a problem.
9.7-2 A decision needs to be made at a decision node.
9.7-3 A random event will occur at a event node.
9.7-4 The probabilities of random events and the payoffs need to be inserted before beginning
analysis.
9.7-5 When performing the analysis, start at the right side of the decision tree and move left one
column at a time.
9.7-6 For each event node, calculate its expected payoff by multiplying the payoff of each
branch by the probability of that branch and then summing these products.
9.7-7 For each decision node, compare the expected payoffs of its branches and choose the
alternative whose branch has the largest expected payoff.
9.7-8 The expected value of sample information (EVSI) is the change in expected value that is
obtained when sample information is obtained. If the EVSI is greater than the cost of
obtaining the information, then it would be worthwhile to obtain the information.
9.8-1 Consolidate the data and results into one section of the spreadsheet.
9.8-2 Performing sensitivity analysis on a piece of data should require changing a value in only
one place on the spreadsheet.
9.8-3 A data table can consider changes in only one or two data cells.
9.8-4 One.
9.8-5 Yes. The spider graph can consider changes in many data cells at a time.
9.8-6 SensIt’s spider graph assumes that each data value varies by the same amount. Sensit’s
tornado diagram overcomes this limitation.
9.9-1 Utilities are intended to reflect the true value of an outcome to the decision-maker.
Chapter 9 - Page 3
Chapter 9
DECISION ANALYSIS
9.9-2
U(M) U(M) U(M)
M M M
Risk averse Risk seeker Risk neutral
9.9-3 Under the assumptions of utility theory, the decision-maker’s utility function for money
has the property that the decision-maker is indifferent between two alternative courses of
action if the two alternatives have the same expected utility.
9.9-4 The decision-maker is offered a lottery with the maximum payoff with probability p and
the minimum payoff with probability 1–p, as compared to a definite payoff of M.
9.9-5 The point of indifference is the value of p where the decision-maker is indifferent between
the two hypothetical alternatives.
9.9-6 The value obtained to evaluate each node of the tree is the expected utility.
9.9-7 Max decided to do the seismic survey and to sell if the result is unfavorable or drill if the
result is favorable.
9.10-1 The Goferbroke problem contained the same elements as typical applications of decision
analysis but is oversimplified.
9.10-2 An influence diagram complements the decision tree for representing and analyzing
decision analysis problems.
9.10-3 Typical participants include management, an analyst, and a group facilitator.
9.10-4 A manager can go to a management consulting firm that specializes in decision analysis.
Chapter 9 - Page 4
Chapter 9
DECISION ANALYSIS
Problems
9.1 a) Max(A1) = 6, Max(A2) = 4, Max(A3) = 8. Maximax = 8 with alternative A3.
b) Min(A1) = 2, Min(A2) = 3, Min(A3) = 1. Maximin = 3 with alternative A2.
9.2 a) Max(A1) = 30, Max(A2) = 31, Max(A3) = 22, Max(A4) = 29. Maximax = 31 with A2.
b) Min(A1) = 20, Min(A2) = 14, Min(A3) = 22, Min(A4) = 21. Maximin = 22 with A3.
Chapter 9 - Page 5
Chapter 9
DECISION ANALYSIS
9.3 a)
State of Nature
Alternative Sell 10 Sell 11 cases Sell 12 Sell 13
cases cases cases
Buy 10 cases $50 $50 $50 $50
Buy 11 cases $47 $55 $55 $55
Buy 12 cases $44 $52 $60 $60
Buy 13 cases $41 $49 $57 $65
Prior 0.2 0.4 0.3 0.1
Probability
b) Max(Buy 10) = $50, Max(Buy 11) = $55, Max(Buy 12) = $60, Max(Buy 13) = $65.
Maximax = $65 with buying 13 cases.
c) Min(Buy 10) = $50, Min(Buy 11) = $47, Min(Buy 12) = $44, Min(Buy 13) = $41.
Maximin = $50 with buying 10 cases.
d) The most likely state of nature is to sell 11 cases. Under this state, she should buy 11
cases with a payoff of $55.
e)
A B C D E F G H
1 Purchase Price $3
2 Selling Price $8
3
4 Payoff Table State of Nature
5 Sell Sell Sell Sell Expected
6 Alternative 10 11 12 13 Payoff
7 Buy 10 $50 $50 $50 $50 $50.00
8 Buy 11 $47 $55 $55 $55 $53.40
9 Buy 12 $44 $52 $60 $60 $53.60
10 Buy 13 $41 $49 $57 $65 $51.40
11
12 Prior Probability 0.2 0.4 0.3 0.1
Jean should buy 12 cases. The maximum expected payoff is $53.60.
Chapter 9 - Page 6
Chapter 9
DECISION ANALYSIS
f) (i) 0.2 and 0.5

A B C D E F G H
1 Purchase Price $3
2 Selling Price $8
3
7 Buy 10 $50 $50 $50 $50 $50.00
8 Buy 11 $47 $55 $55 $55 $53.40
9 Buy 12 $44 $52 $60 $60 $55.20
10 Buy 13 $41 $49 $57 $65 $53.00
11
Jean should purchase 12 cases. The maximum expected payoff is $55.20.

(ii) 0.3 and 0.4
A B C D E F G H
1 Purchase Price $3
2 Selling Price $8
3
7 Buy 10 $50 $50 $50 $50 $50.00
8 Buy 11 $47 $55 $55 $55 $53.40
9 Buy 12 $44 $52 $60 $60 $54.40
10 Buy 13 $41 $49 $57 $65 $52.20
11

(iii) 0.5 and 0.2
A B C D E F G H
1 Purchase Price $3
2 Selling Price $8
3
7 Buy 10 $50 $50 $50 $50 $50.00
8 Buy 11 $47 $55 $55 $55 $53.40
9 Buy 12 $44 $52 $60 $60 $52.80
10 Buy 13 $41 $49 $57 $65 $50.60
11
Chapter 9 - Page 7
Chapter 9
DECISION ANALYSIS
9.4 a) Max(Conservative) = $30 million

Max(Speculative) = $40 million
Max(Countercyclical) = $15 million
Maximax = $40 million with the speculative investment
b) Min(Conservative) = –$10 million
Min(Speculative) = –$30 million
Min(Countercyclical) = –$10 million
Maximin = –$10 million with either the conservative of countercyclical investment.
c) The stable economy is the most likely state of nature.
The speculative investment has the maximum payoff for this state ($10 million).
d) The countercyclical investment has the maximum expected payoff of $5 million.
A B C D E F
1 Payoff Table ($million) Expected
2 Alternative State of Nature (Economy) Payoff
3 (Investment) Improving Stable Worsening ($million)
4 Conservative 30 5 -10 1.5
5 Speculative 40 10 -30 -3
6 Countercyclical -10 0 15 5
7
8 Prior Probability 0.1 0.5 0.4
Chapter 9 - Page 8
Chapter 9
DECISION ANALYSIS
9.5 a) The countercyclical investment has the maximum expected payoff of $8 million.
A B C D E F
4 Conservative 30 5 -10 -1.5
5 Speculative 40 10 -30 -11
7
b) The speculative investment has the maximum expected payoff of $5 million.
A B C D E F
4 Conservative 30 5 -10 4.5
5 Speculative 40 10 -30 5
7
Chapter 9 - Page 9
Chapter 9
DECISION ANALYSIS
c&d)
0.1
Improving
30
30 30
0.7
Conservative Stable
5
0 4.5 5 5
0.2
Worsening
-10
-10 -10
0.1
Improving
40
40 40
0.7
Speculative Stable
2 10
5 0 5 10 10
0.2
Worsening
-30
-30 -30
0.1
Improving
-10
-10 -10
0.7
Counter-cyclical Stable
0
0 2 0 0
0.2
Worsening
15
15 15
Chapter 9 - Page 10
Chapter 9
DECISION ANALYSIS
e)
A B C D E F G H I J K L M N O P
1 0.1
2 Improving Payoff Table ($million)
3 30 Alternative State of Nature (Economy)
4 30 30 (Investment) Improving Stable Worsening
5 Conservative 30 5 -10
6 0.5 Speculative 40 10 -30
7 Conservative Stable Countercyclical -10 0 15
8 5
9 0 1.5 5 5 Prior Probability 0.1 0.5 0.4
10
11 0.4 Investment Countercyclical
12 Worsening
13 -10 Expected Payoff 5
14 -10 -10 ($million)
15
16 0.1
17 Improving
18 40
19 40 40
20
21 0.5
22 Speculative Stable
23 3 10
24 5 0 -3 10 10
25
26 0.4
27 Worsening
28 -30
29 -30 -30
30
31 0.1
32 Improving
33 -10
34 -10 -10
35
36 0.5
37 Countercyclical Stable
38 0
39 0 5 0 0
40
41 0.4
42 Worsening
43 15
44 15 15
Chapter 9 - Page 11
Chapter 9
DECISION ANALYSIS
f) Part a)
M N O P
2 Payoff Table ($million)
3 Alternative State of Nature (Economy)
4 (Investment) Improving Stable Worsening
6 Speculative 40 10 -30
7 Countercyclical -10 0 15
8
10
11 Investment Countercyclical
12
13 Expected Payoff 8
14 ($million)
Part b)
M N O P
2 Payoff Table ($million)
3 Alternative State of Nature (Economy)
4 (Investment) Improving Stable Worsening
6 Speculative 40 10 -30
7 Countercyclical -10 0 15
8
10
11 Investment Speculative
12
13 Expected Payoff 5
14 ($million)
g)
M N O
16 Expected
17 Probability of Optimal Payoff
18 Stable Economy Investment ($million)
19 Speculative 5
20 0 Countercyclical 12.5
21 0.1 Countercyclical 11
22 0.2 Countercyclical 9.5
27 0.7 Speculative 5
Chapter 9 - Page 12
Chapter 9
DECISION ANALYSIS
h)
15
10
0
0 0.2 0.4 0.6 0.8 1 Conservative
Expected Payoff
-5 Speculative
($million)
Countercyclical
-10
-15
-20
-25
Probability of Stable Economy
Counter-cyclical and conservative cross at approximately p=0.62.

Conservative and speculative cross at approximately p = 0.68.
9.6 a) Max(A1) = 80, Max(A2) = 50, Max(A3) = 60.
Maximax = $80 thousand when choosing alternative A1.
b) Min(A1) = 25, Min(A2) = 30, Min(A3) = 40.
Maximin = $40 thousand when choosing alternative A3.
c) S2 is the most likely outcome. For this state, the maximum payoff of $50 thousand
occurs with alternative A2.
d) Alternative A3 has the highest expected payoff of $48 thousand.
A B C D E
1 Payoff Table ($thousand) Expected
2 State of Nature Payoff
3 Alternative S1 S2 ($thousand)
4 A1 80 25 47
5 A2 30 50 42
6 A3 60 40 48
7
8 Prior Probability 0.4 0.6
Chapter 9 - Page 13
Chapter 9
DECISION ANALYSIS
e)
0.4
State 1
80
Alternative 1 80 80
0 47 0.6
State 2
25
25 25
0.4
State 1
30
Alternative 2 30 30
3
48 0 42 0.6
State 2
50
50 50
0.4
State 1
60
Alternative 3 60 60
0 48 0.6
State 2
40
40 40
Chapter 9 - Page 14
Chapter 9
DECISION ANALYSIS
f) When the prior probability of S1 is 0.2, alternative A2 should be chosen, with an

expected payoff of $46 thousand.
A B C D E F G H I J K L M N O
1 0.2
2 State 1 Payoff Table ($thousand)
3 80 State of Nature
4 Alternative 1 80 80 Alternative S1 S2
5 A1 80 25
6 0 36 0.8 A2 30 50
7 State 2 A3 60 40
8 25
9 25 25 Prior Probability 0.2 0.8
10
11 0.2 Best Alternative 2
12 State 1
13 30 Expected Payoff 46
14 Alternative 2 30 30 ($thousand)
15 2
16 46 0 46 0.8
17 State 2
18 50
19 50 50
20
21 0.2
22 State 1
23 60
24 Alternative 3 60 60
25
26 0 44 0.8
27 State 2
28 40
29 40 40
When the prior probability of S1 is 0.6, alternative A1 should be chosen, with an

expected payoff of $58 thousand.
A B C D E F G H I J K L M N O
1 0.6
2 State 1 Payoff Table ($thousand)
3 80 State of Nature
4 Alternative 1 80 80 Alternative S1 S2
5 A1 80 25
6 0 58 0.4 A2 30 50
7 State 2 A3 60 40
8 25
9 25 25 Prior Probability 0.6 0.4
10
11 0.6 Best Alternative 1
12 State 1
13 30 Expected Payoff 58
14 Alternative 2 30 30 ($thousand)
15 1
16 58 0 38 0.4
17 State 2
18 50
19 50 50
20
21 0.6
22 State 1
23 60
24 Alternative 3 60 60
25
26 0 52 0.4
27 State 2
28 40
29 40 40
Chapter 9 - Page 15
Chapter 9
DECISION ANALYSIS
g)
M N O
17 Prior Expected
18 Probability Best Payoff
19 of S1 Alternative ($thousand)
20 2 46
21 0.20 2 46
22 0.24 2 45.2
23 0.28 3 45.6
24 0.32 3 46.4
25 0.36 3 47.2
26 0.40 3 48
27 0.44 1 49.2
28 0.48 1 51.4
29 0.52 1 53.6
30 0.56 1 55.8
31 0.60 1 58
Chapter 9 - Page 16
Chapter 9
DECISION ANALYSIS
9.7 a) Max(A1) = $220 thousand, Max(A2) = $200 thousand.

Maximax = $220 thousand when choosing alternative A1.
b) Min(A1) = $110 thousand, Min(A2) = $150 thousand.
Maximin = $150 thousand when choosing alternative A2.
c) S1 is the most likely outcome. For this state, the maximum payoff of $220 thousand
occurs with alternative A1.
d) Alternative A1 has the highest expected payoff of $194 thousand.
A B C D E F
3 Alternative S1 S2 S3 ($thousand)
4 A1 220 170 110 194
5 A2 200 180 150 189
6
e & f)
0.6
State 1
220
220 220
0.3
Alternative 1 State 2
170
0 194 170 170
0.1
State 3
110
110 110
1
194 0.6
State 1
200
200 200
0.3
Alternative 2 State 2
180
0 189 180 180
0.1
State 3
150
150 150
Chapter 9 - Page 17
Chapter 9
DECISION ANALYSIS
g)
M N O
16 Prior Expected
19 1 194
20 0.30 2 183
21 0.35 2 184
22 0.40 2 185
23 0.45 1 186.5
24 0.50 1 189
25 0.55 1 191.5
26 0.60 1 194
27 0.65 1 196.5
28 0.70 1 199
Let p = prior probability of S1. A1 and A2 cross when p is between 0.40 and 0.45. A
little experimentation reveals that the best alternative changes at p=0.433. They should
choose A2 when p ≤ 0.433, A1 when p > 0.433.
h)
M N O
16 Prior Expected
19 1 194
20 0.30 2 174
21 0.35 2 176.5
22 0.40 2 179
23 0.45 2 181.5
24 0.50 2 184
25 0.55 1 188.5
26 0.60 1 194
27 0.65 1 199.5
28 0.70 1 205
Chapter 9 - Page 18
Chapter 9
DECISION ANALYSIS
i)
M N O
16 Prior Expected
19 1 194
20 0.00 2 180
21 0.05 2 181.5
22 0.10 2 183
23 0.15 1 185
24 0.20 1 188
25 0.25 1 191
26 0.30 1 194
27 0.35 1 197
28 0.40 1 200
j) Alternative A1 should be chosen.
Chapter 9 - Page 19
Chapter 9
DECISION ANALYSIS
9.8 a)
State of Nature (Weather)
Alternative Dry Moderate Damp
Crop 1 20 35 40
Crop 2 22.5 30 45
Crop 3 30 25 25
Crop 4 20 20 20
Prior 0.3 0.5 0.2
Probability
Chapter 9 - Page 20
Chapter 9
DECISION ANALYSIS
b) Crop 1 has the highest expected payoff of $31,500.

0.3
Dry
20
20 20
0.5
Crop 1 Moderate
35
0 31.5 35 35
0.2
Damp
40
40 40
0.3
Dry
22.5
22.5 22.5
0.5
Crop 2 Moderate
30
0 30.75 30 30
0.2
Damp
45
45 45
1
31.5 0.3
Dry
30
30 30
0.5
Crop 3 Moderate
25
0 26.5 25 25
0.2
Damp
25
25 25
0.3
Dry
20
20 20
0.5
Crop 4 Moderate
20
0 20 20 20
0.2
Damp
20
20 20
Chapter 9 - Page 21
Chapter 9
DECISION ANALYSIS
c) When the prior probability of moderate weather is 0.2, Crop 2 has the highest expected
payoff of $35,250.
A B C D E F
2 State of Nature (Weather) Payoff
3 Alternative Dry Moderate Damp ($thousand)
4 Crop 1 20 35 40 33
5 Crop 2 22.5 30 45 35.25
6 Crop 3 30 25 25 26.5
7 Crop 4 20 20 20 20
8
When the prior probability of moderate weather is 0.3, Crop 2 has the highest expected
payoff of $33,750.
A B C D E F
4 Crop 1 20 35 40 32.5
5 Crop 2 22.5 30 45 33.75
6 Crop 3 30 25 25 26.5
7 Crop 4 20 20 20 20
8
payoff of $32,250.
A B C D E F
4 Crop 1 20 35 40 32
5 Crop 2 22.5 30 45 32.25
6 Crop 3 30 25 25 26.5
7 Crop 4 20 20 20 20
8
payoff of $31,000.
A B C D E F
4 Crop 1 20 35 40 31
5 Crop 2 22.5 30 45 29.25
6 Crop 3 30 25 25 26.5
7 Crop 4 20 20 20 20
8
Chapter 9 - Page 22
Chapter 9
DECISION ANALYSIS
9.9 When x = 50, alternative A3 has the highest expected payoff of $5,600.
A B C D E F
1 x= 50
2
3 Payoff Table ($hundred) Expected
5 Alternative S1 S2 S3 ($hundred)
6 A1 100 50 10 54
7 A2 25 40 90 54
8 A3 35 150 30 56
9
When x = 75, alternative A1 has the highest expected payoff of $7,400.
A B C D E F
1 x= 75
2
3 Payoff Table ($hundred) Expected
5 Alternative S1 S2 S3 ($hundred)
6 A1 150 50 10 74
7 A2 25 40 90 54
8 A3 35 225 30 71
9
Barbara Miller should pay a maximum of $1,800 to increase x to 75.
Chapter 9 - Page 23
Chapter 9
DECISION ANALYSIS
9.10 a) Alternative A2 has the highest expected payoff of $1,000.

A B C D E F
1 Payoff Table ($thousands) Expected
3 Alternative S1 S2 S3 ($thousands)
4 A1 4 0 0 0.8
5 A2 0 2 0 1
6 A3 3 0 1 0.9
7
b) With perfect information, choose A1 for when the state is S1, A2 when the state is S2,
and A3 when the state is S3.
EP(with perfect information) = (0.2)(4) + (0.5)(2) + (0.3)(1) = $2,100
EVPI = EP(with perfect information) – EP (without more information)

= $2,100 – $1,000 = $1,100.
Chapter 9 - Page 24
Chapter 9
DECISION ANALYSIS
c)
Alternative 1
4
4 4
0.2
State 1 Alternative 2
1 0
0 4 0 0
Alternative 3
3
3 3
Alternative 1
0
0 0
0.5
2 2
2.1 0 2 2 2
Alternative 3
0
0 0
Alternative 1
0
0 0
0.3
3 0
0 1 0 0
Alternative 3
1
1 1

= $2,100 – $1,000 = $1,100.
d) Since the information will cost $1,000 and the value is no more than $1,100, it might
be worthwhile to spend the money.
Chapter 9 - Page 25
Chapter 9
DECISION ANALYSIS
9.11 a) Alternative A1 has the highest expected payoff of $35.

A B C D E F
1 Payoff Table State of Nature Expected
2 Alternative S1 S2 S3 Payoff
3 A1 $50 $100 -$100 $35
4 A2 $0 $10 -$10 $1
5 A3 $20 $40 -$40 $14
6
b) With perfect information, choose A1 for when the state is S1, A1 when the state is S2,
EP(with perfect information) = (0.5)($50) + (0.3)($100) + (0.2)(–$10) = $53

= $53 – $35 = $18
Chapter 9 - Page 26
Chapter 9
DECISION ANALYSIS
c)
Alternative 1
50
50 50
0.5
1 0
0 50 0 0
Alternative 3
20
20 20
Alternative 1
100
100 100
0.3
1 10
53 0 100 10 10
Alternative 3
40
40 40
Alternative 1
-100
-100 -100
0.2
2 -10
0 -10 -10 -10
Alternative 3
-40
-40 -40

= $53 – $35 = $18
d) Betsy should consider spending up to $18 to obtain more information.
Chapter 9 - Page 27
Chapter 9
DECISION ANALYSIS
9.12 a) Alternative A3 has the highest expected payoff of $35,000.

A B C D E F
3 Alternative S1 S2 S3 ($thousands)
4 A1 -100 10 100 33
5 A2 -10 20 50 29
6 A3 10 10 60 35
7
b) If S1 occurs for certain then choose alternative A3 (payoff is $10,000).
If S1 does not occur for certain then the chance of S2 occurring is 3/8 and the chance of
S3 occurring is 5/8. So choose A1 (expected payoff is $66,250).
A 1: (3/8)(10) + (5/8)(100) = 66.25
A 2: (3/8)(20) + (5/8)(50) = 38.75
A 3: (3/8)(10) + (5/8)(60) = 41.25
EP(with information) = (0.2)(10) + (0.8)(66.25) = 55

EVI = EP (with information) – EP (without more information)
= 55 – 35 = $20,000
The maximum amount you should pay for the information is $20,000.
The decision with this information would be to choose A3 if S1 will occur. Otherwise
choose A1. The expected payoff is $55,000 (excluding the payment for information).
c) If S2 occurs for certain then choose alternative A2 (payoff is $20,000).
A 1: (2/7)(–100) + (5/7)(100) = 42.857
A 2: (2/7)(–10) + (5/7)(50) = 32.857
A 3: (2/7)(10) + (5/7)(60) = 45.714
EP(with information) = (0.3)(20) + (0.7)(42.857) = 38

= 38 – 35 = $3,000
Chapter 9 - Page 28
Chapter 9
DECISION ANALYSIS
d) If S3 occurs for certain then choose alternative A1 (payoff is $100,000).
A1 : (2/5)(–100) + (3/5)(10) = –34
A2 : (2/5)(–10) + (3/5)(20) = 8
A3 : (2/5)(10) + (3/5)(10) = 10
EP(with information) = (0.5)(100) + (0.5)(10) = 55

= 55 – 35 = $20,000
e) With perfect information, choose A3 for when the state is S1, A2 when the state is S2,

= 58 – 35 = $23,000
A maximum of $23,000 should be paid for the information. With perfect information,
choose A3 for when the state is S1, A2 when the state is S2, and A1 when the state is S3.
The resulting expected payoff is $58,000.
f) The maximum amount you should ever pay for testing is $23,000.
Chapter 9 - Page 29
Chapter 9
DECISION ANALYSIS
9.13 a)
Prior Conditional Joint Posterior
Probabilities Probabilities Probabilities Probabilities
P (state) P(finding | state) P(state and finding) P(state | finding)
0.2 0.4
0.8 Oil and FSS Oil, given FSS
FSS, given Oil
USS, given Oil
0.2 0.05 0.1
0.25
Oil Oil and USS Oil, given USS
Dry 0.3 0.6

0.75
0.4 Dry and FSS Dry, given FSS
FSS, given Dry
USS, given Dry
0.6 0.45 0.9
Dry and USS Dry, given USS
b)
B C D E F G H
3 Data: P(Finding | State)
4 State of Prior Finding
5 Nature Probability FSS USS
6 Oil 0.25 0.8 0.2
7 Dry 0.75 0.4 0.6
8
9
10
11
12 Posterior P(State | Finding)
13 Probabilities: State of Nature
14 Finding P(Finding) Oil Dry
15 FSS 0.5 0.4 0.6
16 USS 0.5 0.1 0.9
17
18
19
Chapter 9 - Page 30
Chapter 9
DECISION ANALYSIS
c & d) The optimal policy is to do a seismic survey and sell if it is unfavorable or drill if
it is favorable.
0.1
Oil
670
Drill 800 670
-100 -50 0.9

0.5 Dry
Unfavorable -130
2 0 -130
0 60
Sell
60
90 60
Do seismic survey
0.4
-30 125 Oil
670
Drill 800 670
-100 190 0.6

0.5 Dry
Favorable -130
1 0 -130
0 190
1 Sell
125 60
90 60
0.25
Oil
700
Drill 800 700
-100 100 0.75

Dry
No seismic survey -100
1 0 -100
0 100
Sell
90
90 90
Chapter 9 - Page 31
Chapter 9
DECISION ANALYSIS
9.14 a)
U V W X Y Z AA
3 Data Low Base High
4 Cost of Survey 30 28 30 32
5 Cost of Drilling 100 75 100 140
6 Revenue if Oil 800 600 800 1000
7 Revenue if Sell 90 85 90 95
8 Revenue if Dry 0
9 Prior Probability Of Oil 0.25
10 P(FSS|Oil) 0.6
11 P(USS|Dry) 0.8
12
13
14 Action
15 Do Survey? Yes
16
17 If No If Yes
18
19 Drill Drill If Favorable
20 Sell If Unfavorable
21
22
23
24 Expected Payoff
25 ($thousands)
26 125
27
28
29
33 Oil 0.25 0.8 0.2
34 Dry 0.75 0.4 0.6
35
36
37
38
42 FSS 0.5 0.4 0.6
43 USS 0.5 0.1 0.9
44
45
46
Chapter 9 - Page 32
Chapter 9
DECISION ANALYSIS
b)
Sensit - Sensitivity Analysis - Plot
160
150
140
Expected
130
Payoff
120
110
100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P(FSS|Oil)
190
180
170
160
Expected 150
Payoff 140
130
120
110
100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P(FSS|Dry)
Chapter 9 - Page 33
Chapter 9
DECISION ANALYSIS
c)
Sensit - Sensitivity Analysis - Spider
145
140
135
130 Cost of Survey
Expected Cost of Drilling
Payoff 125
Value Revenue if Oil
120 Revenue if Sell
115
110
105
90% 94% 98% 102% 106% 110%
% Change in Input Value
Sensit - Sensitivity Analysis - Tornado

Revenue if 600
Oil 1000
Cost of Drilling 140 75
Revenue if Sell 85 95
Cost of Survey 32 28
90 100 110 120 130 140 150 160 170

Expected Payoff
Chapter 9 - Page 34
Chapter 9
DECISION ANALYSIS
9.15 a)
State of Nature
Alternative Poor Risk Average Risk Good Risk
Extend Credit -$15,000 $10,000 $20,000
Don’t Extend Credit $0 $0 $0
Prior Probabilities 0.2 0.5 0.3
b) Extending credit maximizes the expected payoff ($8,000).

A B C D E F
2 State of Nature (Credit Record) Payoff
3 Alternative Poor Average Good ($thousands)
4 Extend Credit -15 10 20 8
5 Don't Extend Credit 0 0 0 0
6
c) With perfect information, you would extend credit if their credit record is average or
good, and don’t extend credit if their credit record is poor.

= $11,000 – $8,000 = $3,000.
This indicates that the credit-rating organization should not be used.
Chapter 9 - Page 35
Chapter 9
DECISION ANALYSIS
d) PF = Poor Finding AF = Average Finding GF = Good Finding

PS = Poor State AS = Average State GS = Good State

0.1 0.2778
PF, given PS PS and PF PS, given PF
0.5
0.4 0.08 0.1778
AF, given PS PS and AF PS, given AF
GF, given PS 0.1 0.02 0.1053
PS and GF PS, given GF
0.2 PS
0.2 0.5556
PF, given AS AS and PF AS, given PF
0.4
0.5 0.5 0.25 0.5556
AS AF, given AS AS and AF AS, given AF
GF, given AS 0.1 0.05 0.2632
AS and GF AS, given GF
GS
0.3
0.06 0.1667
PF, given GS GS and PF GS, given PF
0.2
0.4 0.12 0.2667
AF, given GS GS and AF GS, given AF
GF, given GS 0.4 0.12 0.6316
GS and GF GS, given GF
Chapter 9 - Page 36
Chapter 9
DECISION ANALYSIS
e)
A B C D E F G H
1 Template for Posterior Probabilities
2
5 Nature Probability Poor Average Good
6 Poor 0.2 0.5 0.4 0.1
7 Average 0.5 0.4 0.5 0.1
8 Good 0.3 0.2 0.4 0.4
9
10
11
14 Finding P(Finding) Poor Average Good
15 Poor 0.36 0.278 0.556 0.167
16 Average 0.45 0.178 0.556 0.267
17 Good 0.19 0.105 0.263 0.632
18
19
Chapter 9 - Page 37
Chapter 9
DECISION ANALYSIS
f & g) Vincent should not get the credit rating and simply extend credit.
Chapter 9 - Page 38
Chapter 9
DECISION ANALYSIS
0.278
Poor
-20000
-15000 -20000
0.556
Extend Credit Average
5000
0 -290 10000 5000
0.36 0.166
Poor Good
1 15000
0 -290 20000 15000
Don't Extend Credit

-5000
0 -5000
0.178
Poor
-20000
-15000 -20000
0.556
5000
0 3210 10000 5000
0.45 0.266
Get credit rating Average Good
1 15000
-5000 2992.15 0 3210 20000 15000
Don't Extend Credit

-5000
0 -5000
0.105
Poor
-20000
-15000 -20000
0.263
5000
0 8695 10000 5000
0.19 0.632
Good Good
2 1 15000
8000 0 8695 20000 15000
Don't Extend Credit

-5000
0 -5000
0.2
Poor
-15000
-15000 -15000
0.5
10000
0 8000 10000 10000
0.3
Don't get credit rating Good
1 20000
0 8000 20000 20000
Don't Extend Credit

0
0 0
Chapter 9 - Page 39
Chapter 9
DECISION ANALYSIS
h) EVSI = Value with credit rating (ignoring cost) – expected value without
= 8000 – 8000
=0
The sample information provides no value. This is clear because the same decision is
made regardless of the findings of the credit report. (Note there is some rounding error.
Without the rounding error, the expected payoff with the credit rating is 3000 rather
than 2992.15.)
Chapter 9 - Page 40
Chapter 9
DECISION ANALYSIS
9.16 a) Alternative A1 maximizes the expected payoff ($100).

A B C D E
2 Alternative S1 S2 Payoff
3 A1 $400 -$100 $100
4 A2 $0 $100 $60
5
b)
Alternative 1
0.4 400
State 1 400 400
1
0 400
Alternative 2
0
0 0
220
Alternative 1
0.6 -100
State 2 -100 -100
2
0 100
Alternative 2
100
100 100
EVPI = EP (with perfect info) – EP (without more info) = $220 – $100 = $120
This indicates that it might be worthwhile to do the research.

c) P(state and finding) = P(state) P(finding | state)
i) P(Predict S1 and Actual S1) = (0.4)(0.6) = 0.24
ii) P(Predict S1 and Actual S2) = (0.4)(0.4) = 0.16
iii) P(Predict S2 and Actual S1) = (0.6)(0.2) = 0.12
iv) P(Predict S2 and Actual S2) = (0.6)(0.8) = 0.48
d) P(Predict S1) = 0.24 + 0.12 = 0.36
P(Predict S2) = 0.16 + 0.48 = 0.64
e) P(state | finding) = P(state and finding) / P(finding)
P(Actual S1 | Predict S1) = 0.24 / 0.36 = 0.667
Chapter 9 - Page 41
Chapter 9
DECISION ANALYSIS
f)
B C D E F G H
5 Nature Probability Predict S1 Predict S2
6 Actual S1 0.4 0.6 0.4
7 Actual S2 0.6 0.2 0.8
8
9
10
11
14 Finding P(Finding) Actual S1 Actual S2
15 Predict S1 0.36 0.667 0.333
16 Predict S2 0.64 0.250 0.750
17
18
19
g) If S1 is predicted, then choosing alternative A1 maximizes the expected payoff

($233.33).
A B C D E
3 A1 $400 -$100 $233.33
4 A1 $0 $100 $33.33
5
h) If S2 is predicted, then choosing alternative A2 maximizes the expected payoff ($75).
A B C D E
3 A1 $400 -$100 $25
4 A1 $0 $100 $75
5
i) Expected payoff given research is (0.36)($233.33) + (0.64)($75) – $100 = $32.

j) The optimal policy is to do no research and simply choose A1.
Chapter 9 - Page 42
Chapter 9
DECISION ANALYSIS
k)
0.667
S1 occurs
300
Choose A1 400 300
0 133.5 0.333
S2 occurs
0.36 -200
Predict S1 -100 -200
1
0 133.5 0.667
S1 occurs
-100
Choose A2 0 -100
0 -66.7 0.333
S2 occurs
0
Research 100 0
-100 32.06 0.25

S1 occurs
300
Choose A1 400 300
0 -75 0.75
S2 occurs
0.64 -200
Presdict S2 -100 -200
2
0 -25 0.25
S1 occurs
-100
Choose A2 0 -100
2
100 0 -25 0.75
S2 occurs
0
100 0
0.4
S1 occurs
400
Choose A1 400 400
100 0.6
S2 occurs
-100
No Reasearch -100 -100
1
0 100 0.4
S1 occurs
0
Choose A2 0 0
0 60 0.6
S2 occurs
100
100 100
Chapter 9 - Page 43
Chapter 9
DECISION ANALYSIS
9.17 a through d)
0.095 0.6786
0.95 user and positive user, given positive
positive, given user
negative, given user
0.05 0.005 0.0058
0.1
user user and negative user, given negative
nonuser
0.9 0.045 0.3214
0.05 nonuser and positive nonuser, given positive
positive, given nonuser
negative given nonuser
0.95 0.855 0.9942
nonuser and negative nonuser, given negative
e)
B C D E F G H
5 Nature Probability Positive Negative
6 User 0.1 0.95 0.05
7 Nonuser 0.9 0.05 0.95
8
9
10
11
14 Finding P(Finding) User Nonuser
15 Positive 0.14 0.679 0.321
16 Negative 0.86 0.006 0.994
17
18
19
Chapter 9 - Page 44
Chapter 9
DECISION ANALYSIS
9.18 a)
State of Nature
Alternative Successful Unsuccessful
Develop new product $1,500,000 –$1,800,000
Don’t develop new product 0 0
Prior Probabilities 0.667 0.333
b) Choosing to develop the product maximizes the expected payoff ($400,000).

A B C D E
1 Payoff Table ($millions) Expected
3 Alternative Successful Unsuccessful ($millions)
4 Develop Product 1.5 -1.8 0.4
5 Don't Develop Product 0 0 0
6
c) With perfect information, Telemore should develop the product if it would be
successful, and don’t if it will be unsuccessful.
EP(perfect information) = (0.667)(1.5) + (0.333)(0) = $1 million.
EVPI = EP(with perfect information) – EP(without more information)

= $1,000,000 – $400,000 = $600,000.
This indicates that consideration should be given to conducting the market survey.
d)
B C D E F G H
5 Nature Probability Predict Successful Predict Unsuccessful
6 Successful 0.667 0.8 0.2
7 Unsuccessful 0.333 0.3 0.7
8
9
10
11
14 Finding P(Finding) Successful Unsuccessful
15 Predict Successful 0.633 0.842 0.158
16 Predict Unsuccessful 0.367 0.364 0.636
17
18
19
Chapter 9 - Page 45
Chapter 9
DECISION ANALYSIS
e) They should conduct the survey, and develop the product if the survey predicts the
product will be successful. The expected payoff is $520,000.
0.8421
Successful
1.4
Develop Product 1.5 1.4
0 0.8789 0.1579
0.63333 Unsuccessful
Predict Success -1.9
1 -1.8 -1.9
0 0.8789474
Don't Develop
-0.1
0 -0.1
Conduct Survey
0.3636
-0.1 0.52 Successful
1.4
0 -0.7 0.6364
0.36667 Unsuccessful
Predict Unsuccessful -1.9
2 -1.8 -1.9
0 -0.1
1 Don't Develop
0.52 -0.1
0 -0.1
0.6667
Successful
1.5
0 0.4 0.3333
Unsuccessful
No Survey -1.8
1 -1.8 -1.8
0 0.4
Don't Develop
0
0 0
Chapter 9 - Page 46
Chapter 9
DECISION ANALYSIS
f)
0.75
0.7
0.65
0.6
Expected 0.55 Profit if Success
Payoff Loss if Unsuccessful
Value 0.5
Cost of Survey
0.45
0.4
0.35
0.3
75% 80% 85% 90% 95% 100% 105% 110% 115% 120% 125%
Profit if Success1.125 1.875

Loss if Unsuccessful 2.25 1.35
Cost of Survey 0.125 0.075
0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75
Expected Payoff
Chapter 9 - Page 47
Chapter 9
DECISION ANALYSIS
9.19 a)
State of Nature
Alternative p=0.05 p=0.25
Screen –$1,500 –$1,500
Don’t screen –$750 –$3,750
b) Choosing not to screen maximizes the expected payoff. The expected cost is $1,350.
A B C D E
2 Alternative p = 0.05 p = 0.25 Payoff
3 Screen -$1,500 -$1,500 -$1,500
4 Don't Screen -$750 -$3,750 -$1,350
5
c) With perfect information, they would screen if p = 0.25, and don’t screen if p = 0.05.
EP(with perfect information) = (0.8)(–$750) + (0.2)(–$1,500) = –$900
EVPI = EP(with perfect information) – EP(without more information)

= (–$900) – (–$1,350) = $450.
This indicates that consideration should be given to inspecting the single item.
d)
B C D E F G H
5 Nature Probability Defective Nondefective
6 p = 0.05 0.8 0.05 0.95
7 p = 0.25 0.2 0.25 0.75
8
9
10
11
14 Finding P(Finding) p = 0.05 p = 0.25
15 Defective 0.09 0.444 0.556
16 Nondefective 0.91 0.835 0.165
17
18
19
Chapter 9 - Page 48
Chapter 9
DECISION ANALYSIS
e) The optimal policy is not to pre-screen or screen.

0.444
p=0.05
-1625
Screen -1500 -1625
0 -1625 0.556
p=0.25
0.09 -1625
Defective -1500 -1625
1
0 -1625 0.444
p=0.05
-875
Don't screen -750 -875
0 -2543 0.556
p=0.25
-3875
Pre-screen -3750 -3875
-125 -1392.95 0.835

p=0.05
-1625
Screen -1500 -1625
0 -1625 0.165
p=0.25
0.91 -1625
Nondefective -1500 -1625
2
0 -1370 0.835
p=0.05
-875
2
-1350 0 -1370 0.165
p=0.25
-3875
-3750 -3875
0.8
p=0.05
-1500
Screen -1500 -1500
0 -1500 0.2
p=0.25
-1500
Don't Pre-screen -1500 -1500
2
0 -1350 0.8
p=0.05
-750
0 -1350 0.2
p=0.25
-3750
-3750 -3750
f) EVSI = EP(with information ignoring cost of information) – EP(without information)

= (–1392.95 + 125) – (–1350) = (–1267.95) – (–1350) = $82.95.
If the prescreening inspection cost is less than $82.96, then it would be worthwhile to
use.
Chapter 9 - Page 49
Chapter 9
DECISION ANALYSIS
9.20 a)
State of Nature
Alternative Sell 10,000 Sell 100,000
Build Computers $0 $54 million
Sell Rights $15 million $15 million
b)
Sell 10,000
0
Build Computers
Sell 100,000
54
Sell Rights
15
c) They should build computers, with an expected payoff of $27 million.

0.5
Sell 10,000
0
Build Computers 6 0
-6 27 0.5
Sell 100,000
54
1 60 54
27
Sell Rights
15
15 15
Chapter 9 - Page 50
Chapter 9
DECISION ANALYSIS
d)
55
50
45
Expected 40
35
Payoff 30
25
20
15
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Prior Probability of Selling 10,000
e)
54 Crossover
Build point
Expected
Profit
($millions)
15 Sell
0.2 0.4 0.6 0.8 1.0

Prior Probability of Selling 10,000
f) Let p = prior probability of selling 10,000.
For Build:
EP = p(0) + (1 – p)(54)
= –54p + 54
For Sell:
EP = p(15) + (1 – p)(15)
= 15
Build and Sell cross when –54p + 54 = 15 or 54p = 39 or p = 0.722
They should build when p ≤ 0.722, and sell when p > 0.722.
Chapter 9 - Page 51
Chapter 9
DECISION ANALYSIS
9.21 a) With perfect information, they should build computers if they will sell 100,000 of
them, and sell the rights if they could only sell 10,000 computers.
EP(with perfect information) = (0.5)(54) + (0.5)(15) = $34.5 million
EVPI = EP(with perfect information) – EP without more information)

= 34.5 – 27 = $7.5 million.
b) Since the market research will cost $1 million it might be worthwhile to perform it.
c)
0.333 0.667
0.667 sell 10k and sell 10k sell 10k given sell 10k
sell 10k given sell 10k
0.333 0.167 0.333
0.5
sell 10k sell 10k and sell 100k sell 10k given sell 100k
sell 100k
0.5 0.167 0.333
0.333 sell 100k and sell 10k sell 100k given sell 10k
0.667 0.333 0.667
sell 100k and sell 100k sell 100k given sell 100k
Chapter 9 - Page 52
Chapter 9
DECISION ANALYSIS
d)
B C D E F G H
5 Nature Probability Predict Sell 10,000 Predict Sell 100,000
6 Sell 10,000 0.5 0.667 0.333
7 Sell 100,000 0.5 0.333 0.667
8
9
10
11
14 Finding P(Finding) Sell 10,000 Sell 100,000
15 Predict Sell 10,000 0.5 0.667 0.333
16 Predict Sell 100,000 0.5 0.333 0.667
17
18
19
Chapter 9 - Page 53
Chapter 9
DECISION ANALYSIS
9.22 a) The optimal policy is to do no market research and build the computers. The expected
payoff is $27 million.
0.6667
Sell 10,000
-1
Build computers 6 -1
-6 17 0.3333
0.5 Sell 100,000
Predict sell 10,000 53
1 60 53
0 17
Sell rights
14
15 14
Market research
0.3333
-1 26 Sell 10,000
-1
Build computers 6 -1
-6 35 0.6667
0.5 Sell 100,000
Predict sell 100,000 53
1 60 53
0 35
2 Sell rights
27 14
15 14
0.5
Sell 10,000
0
Build computers 6 0
-6 27 0.5
Sell 100,000
No market research 54
1 60 54
0 27
Sell rights
15
15 15
b) EVSI = EP(with information) – EP(without information)

= 27 – 27 = 0.
The information has no value. This is easy to see, as the same decision is made
regardless of the prediction of the market research.
Chapter 9 - Page 54
Chapter 9
DECISION ANALYSIS
c) If the rights can be sold for $16.5 or $13.5 million, the optimal policy is still to build
the computers with an expected payoff of $27 million.
If the cost of setting up the assembly line is $5.4 million or $6.6 million, the optimal
policy is still to build the computers with an expected payoff of $27.6 or $26.4 million,
respectively.
If the difference between the selling price and variable cost of each computer is $540
or $660, the optimal policy is still to build the computers with an expected payoff of
$23.7 or $33.3 million, respectively.
For each combination of financial data, the expected payoff is as shown below. In all
cases, the optimal policy is to build the computers (without market research).
Cost of Selling Price – Expected

Sell Rights Assembly Line Variable Cost Payoff
$13.5 million $5.4 million $540 $24.3 million
Chapter 9 - Page 55
Chapter 9
DECISION ANALYSIS
d)
30.0
25.0
20.0
Expected
15.0
Payoff
10.0
5.0
0.0
13.5 14 14.5 15 15.5 16 16.5
Sell Rights Payoff
28.0
27.5
Expected
27.0
Payoff
26.5
26.0
5.4 5.5 5.6 5.7 5.8 5.9 6 6.1 6.2 6.3 6.4 6.5 6.6
Assembly Line Cost
31.0
30.0
29.0
Expected 28.0
27.0
Payoff 26.0
25.0
24.0
23.0
0.00054 0.00056 0.00058 0.00060 0.00062 0.00064 0.00066
Marginal Profit
Chapter 9 - Page 56
Chapter 9
DECISION ANALYSIS
e)
31.0
30.0
29.0
Expected 28.0 Sell Rights Payoff
Payoff 27.0 Assembly Line Cost
Value 26.0 Marginal Profit
25.0
24.0
23.0
90% 95% 100% 105% 110%
Marginal Profit 0.00054 0.00066

Assembly Line Cost 6.6 5.4
Sell Rights Payoff 13.5
16.5
23.0 24.0 25.0 26.0 27.0 28.0 29.0 30.0 31.0

Expected Payoff
Chapter 9 - Page 57
Chapter 9
DECISION ANALYSIS
9.23 a and b)
0.4
2500
2500
580 0.6
0.2
-700
2 -700
900
900
820 900
0.8
1 800
820 800
750
750
9.24
10
0.5 10
2 0.5
15
0
0
15 0.5
2.5 30
30
0.5
-10
-10
2
8 -5
0.4 -5
2 0.3
5
40
40
5 0.7
8 -10
-10
0.6
10
10
Chapter 9 - Page 58
Chapter 9
DECISION ANALYSIS
9.25 a)
State of Nature
Alternative Winning Season Losing Season
Hold campaign $3 million –$2 million
Don’t hold campaign 0 0
b) Choosing to hold the campaign maximizes the expected payoff ($1 million).
A B C D E
3 Alternative Winning Season Losing Season ($millions)
4 Hold Campaign 3 -2 1
5 Don't Hold Campaign 0 0 0
6
c) With perfect information, Leland University should hold the campaign if they will have
a winning season and don’t hold the campaign if they will have a losing season.
EP(with perfect information) = (0.6)(3) + (0.4)(0) = $1.8 million
EVPI = EP (with perfect info) – EP (without more info)

= $1.8 million – $1 million = $800,000.
Chapter 9 - Page 59
Chapter 9
DECISION ANALYSIS
d)
0.45 0.818
0.75 win and win win. given win
win, given win
lose, given win
0.25 0.15 0.333
0.6
Win win and lose win, given lose
Lose
0.4 0.1 0.182
0.25 lose and win lose, given win
win, given lose
lose, given lose
0.75 0.3 0.667
lose and lose lose, given lose
e)
B C D E F G H
5 Nature Probability Predict W Predict L
6 Winning Season 0.6 0.75 0.25
7 Losing Season 0.4 0.25 0.75
8
9
10
11
14 Finding P(Finding) Winning Season Losing Season
15 Predict W 0.55 0.818 0.182
16 Predict L 0.45 0.333 0.667
17
18
19
Chapter 9 - Page 60
Chapter 9
DECISION ANALYSIS
f & g) Leland University should hire William. If he predicts a winning season then they
should hold the campaign, if he predicts a losing season then they should not hold the
campaign.
0.8182
Win
2.9
Hold campaign 3 2.9
0 1.99091 0.1818
0.55 Lose
Predict win -2.1
1 -2 -2.1
0 1.99091
Don't hold campaign

-0.1
0 -0.1
Hire William
0.3333
-0.1 1.05 Win
2.9
Hold campaign 3 2.9
0 -0.43333 0.6667
0.45 Lose
Predict lose -2.1
2 -2 -2.1
0 -0.1
1 Don't hold campaign

1.05 -0.1
0 -0.1
0.6
Win
3
Hold campaign 3 3
0 1 0.4
Lose
Don't hire William -2
1 -2 -2
0 1
Don't hold campaign

0
0 0
h) EVSI = EP(with information) – EP(without information)

= 1.15 – 1 = 0.15.
The fee can be as high as $150,000 and the information would still be worthwhile.
Chapter 9 - Page 61
Chapter 9
DECISION ANALYSIS
9.26 a & b) (Note: this decision tree continues on the next page.)
0.7
Growth
144
Stocks Y2 24 144
0 133.2 0.3
Recession
0.7 108
Growth -12 108
1
20 133.2 0.7
Growth
126
Bonds Y2 6 126
0 127.8 0.3
Recession
132
12 132
Stocks Y1 0.2
Growth
100 122.94 108
18 108
0.7
Stocks Y2 Recession
81
0 82.8 -9 81
0.1
Depression
0.3 45
Recession -45 45
2
-10 99 0.2
Growth
94.5
4.5 94.5
0.7
Bonds Y2 Recession
99
0 99 9 99
0.1
1 Depression
122.94 108
18 108
Chapter 9 - Page 62
Chapter 9
DECISION ANALYSIS
0.7
Growth
126
Stocks Y2 21 126
0 116.55 0.3
Recession
0.7 94.5
Growth -10.5 94.5
1
5 116.55 0.7
Growth
110.25
Bonds Y2 5.25 110.25
0 111.825 0.3
Recession
115.5
10.5 115.5
Bonds Y1 0.2
Growth
100 117.885 132
22 132
0.7
Stocks Y2 Recession
99
0 101.2 -11 99
0.1
Depression
0.3 55
Recession -55 55
2
10 121 0.2
Growth
115.5
5.5 115.5
0.7
Bonds Y2 Recession
121
0 121 11 121
0.1
Depression
132
22 132
The comptroller should invest in stocks the first year. If there is growth during the first
year then she should invest in stocks again the second year. If there is a recession
during the first year then she should invest in bonds for the second year. The expected
payoff is $122.94 million.
Chapter 9 - Page 63
Chapter 9
DECISION ANALYSIS
9.27 a & b) The optimal policy is to wait until Wednesday to buy if the price is $9 on
Tuesday. If the price is $10 or $11 on Tuesday then buy on Tuesday.
Buy
900
900 900
0.3 0.4
Close at $9 10% Lower
2 810
0 891 810 810
0.3
Wait Unchanged
900
0 891 900 900
0.3
10% Higher
990
990 990
Buy
1000
1000 1000
0.3 0.2
1 900
1007.3 0 1000 900 900
0.2
Wait Unchanged
1000
0 1040 1000 1000
0.6
10% Higher
1100
1100 1100
Buy
1100
1100 1100
0.4 0.1
1 990
0 1100 990 990
0.2
Wait Unchanged
1100
0 1166 1100 1100
0.7
10% Higher
1210
1210 1210
Chapter 9 - Page 64
Chapter 9
DECISION ANALYSIS
9.28 The optimal policy is to sample the fruit and buy if it is excellent and reject if it is
unsatisfactory.
B C D E F G H
5 Nature Probability Excellent Not Excellent
6 Satisfactory Box 0.9 0.8 0.2
7 Unsatisfactory Box 0.1 0.3 0.7
8
9
10
11
14 Finding P(Finding) Satisfactory Box Unsatisfactory Box
15 Excellent 0.75 0.960 0.040
16 Not Excellent 0.25 0.720 0.280
17
18
19
Chapter 9 - Page 65
Chapter 9
DECISION ANALYSIS
0.96
Satisfactory
200
Buy 200 200
0 152 0.04
0.75 Unsatisfactory
Excellent -1000
1 -1000 -1000
0 152
Reject
0
0 0
Sample
0.72
0 114 Satisfactory
200
Buy 200 200
0 -136 0.28
0.25 Unsatisfactory
Not Excellent -1000
2 -1000 -1000
0 0
1 Reject
114 0
0 0
0.9
Satisfactory
200
Buy 200 200
0 80 0.1
Unsatisfactory
Don't sample -1000
1 -1000 -1000
0 80
Reject
0
0 0
Chapter 9 - Page 66
Chapter 9
DECISION ANALYSIS
9.29 a)
State of Nature
Alternative Successful Unsuccessful
Introduce new product $40 million –$15 million
Don’t introduce new product 0 0
A B C D E
3 Alternative Successful Unsuccessful ($millions)
4 Test Market Approves 40 -15 12.5
5 Test Market Doesn't Approve 0 0 0
6
Choose to introduce the new product (expected payoff is $12.5 million).

b) With perfect information, Morton Ward should introduce the product if it will be
successful, and don’t introduce the product if it won’t.
EP(with perfect information) = (0.5)(40) + (0.5)(0) = $20 million.

EVPI = EP (with perfect info) – EP (without more info) = 20 – 12.5 = $7.5 million.
Chapter 9 - Page 67
Chapter 9
DECISION ANALYSIS
c) The optimal policy is not to test but to introduce the new product. The expected payoff
is $12.5 million.
B C D E F G H
5 Nature Probability Approved Not Approved
6 Successful 0.5 0.8 0.2
7 Unsuccessful 0.5 0.25 0.75
8
9
10
11
14 Finding P(Finding) Successful Unsuccessful
15 Approved 0.525 0.762 0.238
16 Not Approved 0.475 0.211 0.789
17
18
19
Chapter 9 - Page 68
Chapter 9
DECISION ANALYSIS
0.762
Successful
38
Introduce product 40 38
0 24.905 0.238
0.525 Unsuccessful
Approved -17
1 -15 -17
0 24.9048
Don't introduce product

-2
0 -2
Test
0.211
-2 12.125 Successful
38
0 -5.4211 0.789
0.475 Unsuccessful
Not approved -17
2 -15 -17
0 -2
2 Don't introduce product

12.5 -2
0 -2
0.5
Successful
40
0 12.5 0.5
Unsuccessful
Don't test -15
1 -15 -15
0 12.5
Don't introduce product

0
0 0
d) EVSI = EP(with information) – EP(without information)

= 14.125 – 12.5 = 1.625.
The cost of the test market can be as high as $1.625 million and still be worthwhile to
do.
e) If the net profit if successful is only $30 million, then the optimal policy is to conduct
the test market and only introduce the product if the test market approves. The
expected payoff is $8.125 million.
If the net profit if successful is $50 million, then the optimal policy is to skip the test
market and introduce the product, with an expected payoff of $17.5 million.
If the net loss if unsuccessful is only $11.25 million, then the optimal policy is to skip
the test market and introduce the product, with an expected payoff of $14.375 million.
Chapter 9 - Page 69
Chapter 9
DECISION ANALYSIS
If the net loss if unsuccessful is $18.75 million, then the optimal policy is to conduct
the test market and only introduce the product if the test market approves. The
expected payoff is $11.656 million.
Net Profit if Net Loss if Optimal Expected

Successful Unsuccessful Policy Payoff
$30 million $11.25 million Skip Test, Introduce Product $9.375 million
$30 million $18.75 million Test, Introduce if Approve $7.656 million
$50 million $11.25 million Skip Test, Introduce Product $19.375 million
$50 million $18.75 million Test, Introduce if Approve $15.656 million
f)
18
16
Expected 14
Payoff 12
10
8
30 32 34 36 38 40 42 44 46 48 50
Net Profit if Successful
14.5
14
13.5
Expected
13
Payoff
12.5
12
11.5
11 12 13 14 15 16 17 18 19
Net Loss if Unsuccessful
Chapter 9 - Page 70
Chapter 9
DECISION ANALYSIS
g)
18
17
16
15
Expected 14 Net Profit if Successful
Payoff 13
Value 12 Net Loss if Unsuccessful
11
10
9
8
75% 85% 95% 105% 115% 125%
Net Profit if Successful30 50

18.75 11.25
8 9 10 11 12 13 14 15 16 17 18
Expected Payoff
Both charts indicate that the expected profit is sensitive to both parameters, but is
somewhat more sensitive to changes in the profit if successful than to changes in the
loss if unsuccessful.
Chapter 9 - Page 71
Chapter 9
DECISION ANALYSIS
9.30 a) Chelsea should run in the NH primary. If she does well then she should run in the ST
primaries. If she does poorly then she should not run in the ST primaries. The
expected payoff is $666,667.
0.5714
Do well in ST
14.4
Run in ST 16 14.4
0 3.257 0.4286
0.4667 Do poorly in ST
Do well in NH -11.6
1 -10 -11.6
0 3.2571
Don't run in ST
-1.6
0 -1.6
Run in NH
0.25
-1.6 0.66667 Do well in ST
14.4
Run in ST 16 14.4
0 -5.1 0.75
0.5333 Do poorly in ST
Do poorly in NH -11.6
2 -10 -11.6
0 -1.6
1 Don't run in ST
0.66667 -1.6
0 -1.6
0.4
Do well in ST
16
Run in ST 16 16
0 0.4 0.6
Do poorly in ST
Don't run in NH -10
1 -10 -10
0 0.4
Don't run in ST
0
0 0
b) If the payoff for a doing well in ST is only $12 million, Chelsea should not run in
either NH or ST, with an expected payoff of $0.
If the payoff for doing well in ST is $20 million, Chelsea should not run in NH, but
should run in ST, with an expected payoff of $2 million.
If the loss for doing poorly in ST is $7.5 million, Chelsea should not run in NH, but
should run in ST, with an expected payoff of $1.9 million.
If the loss for doing poorly in ST is $12.5 million, Chelsea should run in NH, and then
run in ST if she does well in NH, with an expected payoff of $166,667.
Chapter 9 - Page 72
Chapter 9
DECISION ANALYSIS
Payoff if do Loss if do Optimal Expected

Well in ST Poorly in ST Policy Payoff
$12 million $7.5 million Run in ST only $300,000
$12 million $12.5 million Don’t run in either $0
$20 million $7.5 million Run in ST only $3.5 million
$20 million $12.5 million Run in NH, Run in ST if do well $1.233 million
c)
2.0
1.5
Expected
1.0
Payoff
0.5
0.0
12 13 14 15 16 17 18 19 20
Payoff if Win ST
1.5
Expected
1
Payoff
0.5
0
7 8 9 10 11 12 13
Loss if Lose ST
Chapter 9 - Page 73
Chapter 9
DECISION ANALYSIS
d)
Sensit - Sensitivity Analysis - Spide
2.0
1.5
Expected Payoff if Win ST
Payoff 1.0
Value Loss if Lose ST
0.5
0.0
75% 85% 95% 105% 115% 125%
Payoff if Win ST
12 20
12.5 7.5
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Expected Payoff
Both charts indicate that the expected payoff is sensitive to both parameters, although it
is slightly more sensitive to changes in the profit if she does well than to changes in the
loss if she does poorly.
Chapter 9 - Page 74
Chapter 9
DECISION ANALYSIS
9.31 a)
0.95455
Successful
1960
Hire 2000 1960
0 1841.8 0.04545
0.66 Unsuccessful
Pass -640
1 -600 -640
0 1842
Don't hire
-40
0 -40
Test
0.20588
-40 1202 Successful
1960
Hire 2000 1960
0 -104.7 0.79412
0.34 Unsuccessful
Fail -640
2 -600 -640
0 -40
2 Don't hire
1220 -40
0 -40
0.7
Successful
2000
Hire 2000 2000
0 1220 0.3
Unsuccessful
Don't test -600
1 -600 -600
0 1220
Don't hire
0
0 0
Chapter 9 - Page 75
Chapter 9
DECISION ANALYSIS
B C D E F G H
5 Nature Probability Pass Test Fail Test
6 Successful 0.7 0.9 0.1
7 Not Successful 0.3 0.1 0.9
8
9
10
11
14 Finding P(Finding) Successful Not Successful
15 Pass Test 0.6600 0.9545 0.0455
16 Fail Test 0.3400 0.2059 0.7941
17
18
19
The optimal policy is not to pay for testing and to hire Matthew.
b) If the fee is less than $22,000 then the testing is worthwhile.
Chapter 9 - Page 76
Chapter 9
DECISION ANALYSIS
9.32 a & b) They should do no seismic survey, and sell the land, with an expected utility of
90.
0.143
Oil
440
Drill 440 440
0 -108.48 0.857
0.7 Dry
Unfavorable -200
2 -200 -200
0 60
Sell
60
60 60
Do seismic survey
0.5
0 78 Oil
440
Drill 440 440
0 120 0.5
0.3 Dry
Favorable -200
1 -200 -200
0 120
2 Sell
90 60
60 60
0.25
Oil
450
Drill 450 450
0 15 0.75
Dry
No seismic survey -130
2 -130 -130
0 90
Sell
90
90 90
9.33 a) Choosing not to buy insurance maximizes the expected payoff ($249,840).
A B C D E
2 Alternative Earthquake No Earthquake Payoff
3 Insurance $249,820 $249,820 $249,820
4 No Insurance $90,000 $250,000 $249,840
5
b) U(insurance) = U(250,000-180) = 249,820 = 499.82

U(no insurance) = (0.999)U(250,000) + (0.001)U(90,000) = 499.8
The optimal policy is to buy insurance.
Chapter 9 - Page 77
Chapter 9
DECISION ANALYSIS
9.34 E(utility) of $19,000 = U(19) = 25 = 5

E(utility) of investment = (0.3)U(10) + (0.7)U(30) = (0.3) 16  (0.7) 36 = 5.4
Choose the investment to maximize expected utility.
9.35 U(10) = 0, U(30) = 1.
U(19) = p such that you are indifferent between 30 with probability p and 10 with
probability 1–p, compared to 19 for sure.
The parents are offering p = 0.7. If you are indifferent at this p, then U(19) = 0.7.
9.36 a) U(1) = p = 0.125.
b) E(5) = p = 0.5625.
c) Answers will vary.
Chapter 9 - Page 78
Chapter 9
DECISION ANALYSIS
9.37 a) Expected utility of A1 = pU(25) + (1 – p)U(36) = 5p + 6(1 – p) = 6 – p.

Expected utility of A2 = pU(100) + (1 – p)U(0) = 10p + 0 = 10p.
Expected utility of A3 = pU(0) + (1 – p)U(49) = 7(1 – p) = 7 – 7p.
Expected
Utility
10 A2
7
6
A1
A3
0.2 0.4 0.6 0.8 1

p
A1 and A3 cross when 6 – p = 7 – 7p, or p = 1/6.

A1 and A2 cross when 6 – p =10p, or p = 6/11.
Thus, A3 is best when p ≤ 1/6, A1 is best when 1/6 < p ≤ 6/11, and A2 is best when
p > 6/11.
Chapter 9 - Page 79
Chapter 9
DECISION ANALYSIS
b) When p = 0.25, alternative A1 is optimal with an expected utility of 0.4833.

0.25
State 1
25 p= 0.25
Alternative 1 25 25
0.393469 RT = 50
0 33.015 0.75
0.4833 State 2
36
36 36
0.513248
0.25
State 1
100
Alternative 2 100 100
1 0.864665
33.0149 0 12.178 0.75
0.4833 0.2162 State 2
0
0 0
0
0.25
State 1
0
Alternative 3 0 0
0
0 31.604 0.75
0.4685 State 2
49
49 49
0.6246889
Chapter 9 - Page 80
Chapter 9
DECISION ANALYSIS
When p = 0.5, alternative A1 is optimal with an expected utility of 0.4534.

0.5
State 1
25 p= 0.5
Alternative 1 25 25
0.393469 RT = 50
0 30.198 0.5
0.4534 State 2
36
36 36
0.513248
0.5
State 1
100
1 0.864665
30.1981 0 28.311 0.5
0.45336 0.4323 State 2
0
0 0
0
0.5
State 1
0
Alternative 3 0 0
0
0 18.723 0.5
0.3123 State 2
49
49 49
0.6246889
Chapter 9 - Page 81
Chapter 9
DECISION ANALYSIS
When p = 0.75, alternative A2 is optimal with an expected utility of 0.6485.

0.75
State 1
25 p= 0.75
Alternative 1 25 25
0.393469 RT = 50
0 27.532 0.25
0.4234 State 2
36
36 36
0.513248
0.75
State 1
100
2 0.864665
52.2771 0 52.277 0.25
0.6485 0.6485 State 2
0
0 0
0
0.75
State 1
0
Alternative 3 0 0
0
0 8.4903 0.25
0.1562 State 2
49
49 49
0.6246889
Chapter 9 - Page 82
Chapter 9
DECISION ANALYSIS
9.38 The optimal policy is not to test for disease A but to treat disease A. (Note: this decision
tree is continued on the next page.)
0.5
Poor Health
0.8 7
Have A 10 7
0 17 0.5
Good Health
Treat A 27
30 27
-1 13
0.2 1
Have B Die
-3
0 -3 0 -3
0.5
Positive test result 0.2
1 Die
0 13 0.8 -2
Have A 0 -2
0 6 0.8
Poor Health
8
Don't treat A 10 8
0 5.4 0.5
Die
0.2 -2
Have B 0 -2
0 3 0.5
Poor Health
Test for A 8
10 8
-2 8.3
0.5
Poor Health
0.2 7
Have A 10 7
0 17 0.5
Good Health
Treat A 27
30 27
-1 1
0.8 1
Have B Die
-3
0 -3 0 -3
0.5
Negative test result 0.2
2 Die
0 3.6 0.2 -2
Have A 0 -2
0 6 0.8
Poor Health
8
Don't treat A 10 8
2
9 0 3.6 0.5
Die
0.8 -2
Have B 0 -2
0 3 0.5
Poor Health
8
10 8
Chapter 9 - Page 83
Chapter 9
DECISION ANALYSIS
0.5
Poor Health
0.5 9
Have A 10 9
0 19 0.5
Good Health
Treat A 29
30 29
-1 9
0.5 1
Have B Die
-1
0 -1 0 -1
Don't test for A 0.2

1 Die
0 9 0.5 0
Have A 0 0
0 8 0.8
Poor Health
10
Don't treat A 10 10
0 6.5 0.5
Die
0.5 0
Have B 0 0
0 5 0.5
Poor Health
10
10 10
B C D E F G H
5 Nature Probability Positive Negative
6 Disease A 0.5 0.8 0.2
7 Disease B 0.5 0.2 0.8
8
9
10
11
14 Finding P(Finding) Disease A Disease B
15 Positive 0.5 0.8 0.2
16 Negative 0.5 0.2 0.8
17
18
19
Chapter 9 - Page 84
Chapter 9
DECISION ANALYSIS
9.39 For the decision-maker to be indifferent between A1 and A2, U(x) must equal 3.5. So,
x1/3 = 3.5 or x = $42.88.
A1 $x
x1/3
0.6
$7,600
19
-$600 19
7 0.4
0.5 -$590
$0 -11
1 0 -11
7
-$141
A2 $16
-5
3.5 -5
0.5
$141
0
0
Chapter 9 - Page 85
Chapter 9
DECISION ANALYSIS
9.40 a)
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA
3 0.1429
4 Oil
5 580
6 Drill 580 580
7
8 0 -45.714 0.8571
9 0.7 Dry
10 Unfavorable -150
11 2 -150 -150
12 0 60 Data
13 Prior Prob. Of Oil 0.25
14 Sell P(FSS|Oil) 0.6
15 60 P(USS|Dry) 0.8
16 60 60
17 Do Survey
18 0.5 Action
19 0 106.5 Oil Do Survey? Yes
20 580
21 Drill 580 580 If No If Yes
22
23 0 215 0.5 Sell Drill If Favorable
24 0.3 Dry Sell If Unfavorable
25 Favorable -150
26 1 -150 -150 Expected Utility
27 0 215 106.5
28
29 1 Sell
30 106.5 60 Data: P(Finding | State)
31 60 60 State of Prior Finding
33 0.25 Oil 0.25 0.6 0.4
34 Oil Dry 0.75 0.2 0.8
35 600
36 Drill 600 600
37
38 0 71.25 0.75
39 Dry Posterior P(State | Finding)
40 No Survey -105 Probabilities: State of Nature
41 2 -105 -105 Finding P(Finding) Oil Dry
42 0 90 FSS 0.3 0.5 0.5
43 USS 0.7 0.1429 0.86
44 Sell
45 90
46 90 90
Chapter 9 - Page 86
Chapter 9
DECISION ANALYSIS
b) When the prior probability of oil is 15%, the optimal policy is to do no survey and sell
the land, with an expected utility of 90.
U V W X Y Z AA
12 Data
14 P(FSS|Oil) 0.6
15 P(USS|Dry) 0.8
16
17
18 Action
19 Do Survey? No
20
21 If No If Yes
22
23 Sell Drill If Favorable
25
26 Expected Utility
27 90
28
29
33 Oil 0.15 0.6 0.4
34 Dry 0.85 0.2 0.8
35
36
37
38
42 FSS 0.26 0.3462 0.65
43 USS 0.74 0.0811 0.92
44
45
46
Chapter 9 - Page 87
Chapter 9
DECISION ANALYSIS
When the prior probability of oil is 20%, the optimal policy is to do the survey and
drill if favorable or sell if unfavorable, with an expected utility of 90.86.
U V W X Y Z AA
12 Data
14 P(FSS|Oil) 0.6
15 P(USS|Dry) 0.8
16
17
18 Action
19 Do Survey? Yes
20
21 If No If Yes
22
23 Sell Drill If Favorable
25
26 Expected Utility
27 90.857143
28
29
33 Oil 0.2 0.6 0.4
34 Dry 0.8 0.2 0.8
35
36
37
38
42 FSS 0.28 0.4286 0.57
43 USS 0.72 0.1111 0.89
44
45
46
Chapter 9 - Page 88
Chapter 9
DECISION ANALYSIS
When the prior probability of oil is 30%, the optimal policy is to do the survey and
drill if favorable or sell if unfavorable, with an expected utility of 120.19.
U V W X Y Z AA
12 Data
14 P(FSS|Oil) 0.6
15 P(USS|Dry) 0.8
16
17
18 Action
19 Do Survey? Yes
20
21 If No If Yes
22
25
26 Expected Utility
27 120.1875
28
29
33 Oil 0.3 0.6 0.4
34 Dry 0.7 0.2 0.8
35
36
37
38
42 FSS 0.32 0.5625 0.44
43 USS 0.68 0.1765 0.82
44
45
46
Chapter 9 - Page 89
Chapter 9
DECISION ANALYSIS
When the prior probability of oil is 35%, the optimal policy is to skip the survey and
drill for oil, with an expected utility of 141.75.
U V W X Y Z AA
12 Data
14 P(FSS|Oil) 0.6
15 P(USS|Dry) 0.8
16
17
18 Action
19 Do Survey? No
20
21 If No If Yes
22
25
26 Expected Utility
27 141.75
28
29
33 Oil 0.35 0.6 0.4
34 Dry 0.65 0.2 0.8
35
36
37
38
42 FSS 0.34 0.6176 0.38
43 USS 0.66 0.2121 0.79
44
45
46
Chapter 9 - Page 90
Chapter 9
DECISION ANALYSIS
Cases
9.1 a) The course of action that maximizes the expected payoff is to answer the $500,000
question alone. If you get that question correct, then use the phone-a-friend lifeline to
help answer to $1 million question. The expected payoff is $440,980.
Prob(Correct) 0.5
$1 million (alone) 0.5 Correct
$1 million (Phone-a-Friend) 0.65 $1,000,000
$500k (alone) 0.65 Answer alone 1000000 $1,000,000
$500k (Phone-a-Friend) 0.8
0 $516,000 0.5
0.8 Incorrect
Correct $32,000
1 32000 $32,000
0 $516,000
Answer with
Phone-a-Friend Don't Answer
$500,000
0 $419,200 500000 $500,000
0.2
Incorrect
$32,000
32000 $32,000
0.65
Correct
Answer with $1,000,000
Phone-a-Friend 1000000 $1,000,000
0 $661,200 0.35
0.65 Incorrect
2 Correct $32,000
$440,980 1 32000 $32,000
0 $661,200
Answer alone Don't Answer

$500,000
0 $440,980 500000 $500,000
0.35
Incorrect
$32,000
32000 $32,000
Don't Answer
$250,000
250000 $250,000
b) Answers will vary depending on your level of risk aversion. Here is one possible
solution:
Set U(Maximum) = U($1 million) = 1.
Set U(Minimum) = U($32 thousand) = 0.
If getting $250 thousand for sure is equivalent to a 60% chance of getting $1 million
vs. a 40% chance of getting $32 thousand, then U($250 thousand) = p = 0.6.
If getting $500 thousand for sure is equivalent to a 90% chance of getting $1 million
vs. a 10% chance of getting $32 thousand, then U($250 thousand) = p = 0.9.
Chapter 9 - Page 91
Chapter 9
DECISION ANALYSIS
c) With the utilities derived in part b, the decision changes to using the phone-a-friend
lifeline to help answer the $500 thousand question, and then walk away.
Prob(Correct)
$1 million (alone) 0.5 0.5
$1 million (Phone-a-Friend) 0.65 Correct
$500k (alone) 0.65 1
$500k (Phone-a-Friend) 0.8 Answer alone 0 1 $1,000,000
U($1 million) = 1 0 0.5 0.5

U($500k) = 0.9 0.8 Incorrect
U($250k) = 0.6 Correct 0
U($32k) = 0 2 0 0 $32,000
0 0.9
Answer with
Phone-a-Friend Don't Answer
0.9
0 0.72 0 0.9 $500,000
0.2
Incorrect
0
0 0 $32,000
0.65
Correct
Answer with 1
Phone-a-Friend 0 1 $1,000,000
0 0.65 0.35
0.65 Incorrect
1 Correct 0
0.72 2 0 0 $32,000
0 0.9
Answer alone Don't Answer

0.9
0 0.585 0 0.9 $500,000
0.35
Incorrect
0
0 0 $32,000
Don't Answer
0.6
0 0.6 $250,000
Chapter 9 - Page 92
Chapter 9
DECISION ANALYSIS
9.2 a)
A B C D E F G H
1 Template for Posterior Probabilities
2
5 Nature Probability Well in Test Poor in Test
6 Well in Full Market 0.5 0.8 0.2
7 Poor in Full Market 0.5 0.4 0.6
8
9
10
11
14 Finding P(Finding) Well in Full Market Poor in Full Market
15 Well in Test 0.6 0.666666667 0.333333333
16 Poor in Test 0.4 0.25 0.75
17
18
19
Chapter 9 - Page 93
Chapter 9
DECISION ANALYSIS
b) The best course of action is to skip the test market, and immediately market the product
fully. The expected payoff is $1750.
0.66667
Do Well
Probabilities: $1,300
P("Well in Full Market") 0.5 Fully Market $2,000 $1,300
P("Well in Test") 0.6
P("Well in Full Market | Well in Test"0.6667 -$1,000 $700 0.33333
P("Well in Full Market | Poor in Test") 0.25 0.2 Do Poorly
P("LSPAF") 0.2 LSPAF Enters Market -$500
1 $200 -$500
Cost & Revenue Data: 0 $700
Do Well, LSPAF $2,000
Do Well, No LSPAF $5,000 Don't
Do Poorly, LSPAF $200 $300
Do Poorly, No LSPAF $500 0.60 0 $300
Do Well in Test Market $400 Do Well
Do Poorly in Test Market $40 0.66667
Full Market Cost $1,000 $400 $2,380 Do Well
Test Market Cost $100 $4,300
Fully Market $5,000 4300
-$1,000 $2,800 0.33333

0.8 Do Poorly
No LSPAF -$200
1 $500 -200
0 $2,800
Don't
$300
0 $300
Test Market
0.25
-$100 $1,604 Do Well
$940
Fully Market $2,000 $940
-$1,000 -$410 0.75

0.2 Do Poorly
LSPAF Enters Market -$860
2 $200 -$860
0 -$60
Don't
-$60
0.40 0 -$60
Do Poorly
0.25
$40 $440 Do Well
$3,940
Fully Market $5,000 $3,940
2
$1,750 -1000 565 0.75
Expected Profit 0.8 Do Poorly
No LSPAF -$560
1 $500 -$560
0 $565
Don't
-$60
0 -$60
0.5
Do Well
$4,000
Fully Market ##### $4,000
##### 1750 0.5

Do Poorly
No Test Market -$500
1 $500 -$500
0 $1,750
Don't
0
0 0
Chapter 9 - Page 94
Chapter 9
DECISION ANALYSIS
c) If the probability that the LSPAFs enter the market before the test marketing would be
completed increases this would make the test market even less desirable, so it would
still not be worthwhile to do. However, if the probability decreases, this would make
the test market more desirable. It might reach the point where the test market is
worthwhile.
d)
Expected Test
Prob(LSPAF) Payoff Market?
$1,750 No
0.0 $1,906 Yes
0.1 $1,755 Yes
0.2 $1,750 No
0.3 $1,750 No
0.4 $1,750 No
0.5 $1,750 No
0.6 $1,750 No
0.7 $1,750 No
0.8 $1,750 No
0.9 $1,750 No
1.0 $1,750 No
e) It is better to perform the test market if the probability that the LSPAFs enter the
market is 10% or less. It is better to skip the test market if the probability is greater
than 10%.
Chapter 9 - Page 95
Chapter 9
DECISION ANALYSIS
9.3 a) The decision alternatives are to price the product high ($50), medium ($40), or low
($30), or don’t market the product at all. The possible states of nature are the demand
could be high (50,000), medium (30,000), or low (20,000), which in turn depends upon
the price, and whether the competition is severe, moderate, or weak. The various data
are summarized in the spreadsheet below.
A B C D E F G H I
1 Price Severe Moderate Weak
2 High $50 Prior Probability 0.2 0.7 0.1
3 Medium $40
4 Low $30 Prior Revenue
5 High Price Severe Moderate Weak Probability ($thousands)
6 Sales Sales High 0.20 0.25 0.30 0.245 2,500
7 (thousands) Sales Medium 0.25 0.30 0.35 0.295 1,500
8 High 50 Sales Low 0.55 0.45 0.35 0.46 1,000
9 Medium 30
10 Low 20 Medium Price Severe Moderate Weak
11 Sales High 0.25 0.30 0.40 0.3 2,000
12 Sales Medium 0.35 0.40 0.50 0.4 1,200
13 Sales Low 0.40 0.30 0.10 0.3 800
14
15 Low Price Severe Moderate Weak
16 Sales High 0.35 0.40 0.50 0.4 1,500
17 Sales Medium 0.40 0.50 0.45 0.475 900
18 Sales Low 0.25 0.10 0.05 0.125 600
H I
4 Prior Revenue
5 Probability ($thousands)
6 =SUMPRODUCT($E$2:$G$2,E6:G6) =$B$2*B8
9
10
14
15
The payoff table can be generated based on the results in column I.
Chapter 9 - Page 96
Chapter 9
DECISION ANALYSIS
The decision tree for this problem follows (over three pages):
0.20
Sales High
2,500
2,500 2,500
0.2 0.25
Severe Competition Sales Medium
1,500
0 1425 1,500 1,500
0.55
Sales Low
1,000
1,000 1,000
0.25
Sales High
2,500
2,500 2,500
0.7 0.30
Price High Moderate Competition Sales Medium
1,500
0 1515 0 1525 1,500 1,500
0.45
Sales Low
1,000
1,000 1,000
0.30
Sales High
2,500
2,500 2,500
0.1 0.35
Weak Competition Sales Medium
1,500
0 1625 1,500 1,500
0.35
Sales Low
1,000
1,000 1,000
Chapter 9 - Page 97
Chapter 9
DECISION ANALYSIS
0.25
Sales High
2,000
2,000 2,000
0.2 0.35
1,200
0 1240 1,200 1,200
0.40
Sales Low
800
800 800
0.30
Sales High
2,000
2,000 2,000
0.7 0.40
Price Medium Moderate Competition Sales Medium
1 1,200
1515 0 1320 0 1320 1,200 1,200
0.30
Sales Low
800
800 800
0.40
Sales High
2,000
2,000 2,000
0.1 0.50
1,200
0 1480 1,200 1,200
0.10
Sales Low
800
800 800
Chapter 9 - Page 98
Chapter 9
DECISION ANALYSIS
0.35
Sales High
1,500
1,500 1,500
0.2 0.40
900
0 1035 900 900
0.25
Sales Low
600
600 600
0.40
Sales High
1,500
1,500 1,500
0.7 0.50
Price Low Moderate Competition Sales Medium
900
0 1102.5 0 1110 900 900
0.10
Sales Low
600
600 600
0.50
Sales High
1,500
1,500 1,500
0.1 0.45
900
0 1185 900 900
0.05
Sales Low
600
600 600
Chapter 9 - Page 99
Chapter 9
DECISION ANALYSIS
b) The most likely state depends upon the price. As calculated in the spreadsheet below, if
the price is set high ($50), the most likely outcome is low sales (20,000), with $1
million in revenue. If the price is set medium ($40), the most likely outcome is medium
sales (30,000), with $1.2 million in revenue. If the price is set low ($30), the most
likely outcome is medium sales (30,000), with $0.9 million in revenue.
A B C D E F G H I
3 Medium $40
6 Sales Sales High 0.20 0.25 0.30 0.245 2,500
8 High 50 Sales Low 0.55 0.45 0.35 0.46 1,000
9 Medium 30
11 Sales High 0.25 0.30 0.40 0.3 2,000
12 Sales Medium 0.35 0.40 0.50 0.4 1,200
13 Sales Low 0.40 0.30 0.10 0.3 800
14
16 Sales High 0.35 0.40 0.50 0.4 1,500
17 Sales Medium 0.40 0.50 0.45 0.475 900
18 Sales Low 0.25 0.10 0.05 0.125 600
Thus, to maximize revenue under the maximum likelihood criterion, Charlotte should
charge the medium price ($40), for a most likely outcome of $1.2 million in revenue.
Chapter 9 - Page 100

Chapter 9
DECISION ANALYSIS
c) As shown in the decision tree for part a (recall that decision trees assume Bayes’
decision rule), Charlotte should charge the high price ($50), since this maximizes the
expected revenue ($1.515 million). Alternatively, the expected revenues for each
possible decision can be calculated directly as shown in the following spreadsheet.
A B C D E F G H I J
3 Medium $40 Expected
4 Low $30 Prior Revenue Revenue
5 High Price Severe Moderate Weak Probability ($thousands) ($thousands)
6 Sales Sales High 0.20 0.25 0.30 0.245 2,500
7 (thousands) Sales Medium 0.25 0.30 0.35 0.295 1,500 1,515
8 High 50 Sales Low 0.55 0.45 0.35 0.46 1,000
9 Medium 30
11 Sales High 0.25 0.30 0.40 0.3 2,000
12 Sales Medium 0.35 0.40 0.50 0.4 1,200 1,320
13 Sales Low 0.40 0.30 0.10 0.3 800
14
16 Sales High 0.35 0.40 0.50 0.4 1,500
17 Sales Medium 0.40 0.50 0.45 0.475 900 1,102.5
18 Sales Low 0.25 0.10 0.05 0.125 600
H I J
3 Expected
4 Prior Revenue Revenue
5 Probability ($thousands) ($thousands)
7 =SUMPRODUCT($E$2:$G$2,E7:G7) =$B$2*B9 =SUMPRODUCT(H6:H8,I6:I8)
9
10
14
15

Chapter 9
DECISION ANALYSIS
d) With more information from the marketing research company, the posterior
probabilities for the state of competition can be found using the template for posterior
probabilities as follows.
B C D E F G H
4 Nature Probability Predict Severe Predict Moderate Predict Weak
5 Severe 0.2 0.8 0.15 0.05
6 Moderate 0.7 0.15 0.8 0.05
7 Weak 0.1 0.03 0.07 0.9
8
9
10
13 Finding P(Finding) Severe Moderate Weak
14 Predict Severe 0.268 0.597 0.392 0.011
15 Predict Moderate 0.597 0.050 0.938 0.012
16 Predict Weak 0.135 0.074 0.259 0.667
17
18
To keep the decision tree from becoming too unwieldy, we will break it into parts. The
first three parts consider the situation after each possible prediction by the marketing
research company. The decision tree from part a is reused with the only change being
the prior probabilities of severe, moderate and weak competition used in part a are
replaced by the appropriate posterior probabilities calculated above, depending upon
the prediction of the marketing research company. For example, if the marketing
research company predicts the competition will be severe, the probability of severe,
moderate, and weak competition are 0.597, 0.392, and 0.011, respectively.

Chapter 9
DECISION ANALYSIS
The optimal decision if the marketing research company predicts severe is to price high
($50), with expected revenue of $1.466 million.
A B C D E F G H I
3 Medium $40
6 Sales Sales High 0.20 0.25 0.30 0.220709 2,500
8 High 50 Sales Low 0.55 0.45 0.35 0.5085821 1,000
9 Medium 30
11 Sales High 0.25 0.30 0.40 0.2712687 2,000
12 Sales Medium 0.35 0.40 0.50 0.3712687 1,200
13 Sales Low 0.40 0.30 0.10 0.3574627 800
14
16 Sales High 0.35 0.40 0.50 0.3712687 1,500
17 Sales Medium 0.40 0.50 0.45 0.4397388 900
18 Sales Low 0.25 0.10 0.05 0.1889925 600
19
20 Optimal Decision Price High
21
22 Expected Revenue 1466.42
23 ($thousands)
The optimal decision if the marketing research company predicts moderate competition
is to price high ($50), with expected revenue of $1.521 million.
A B C D E F G H I
3 Medium $40
6 Sales Sales High 0.20 0.25 0.30 0.2480737 2,500
8 High 50 Sales Low 0.55 0.45 0.35 0.4538526 1,000
9 Medium 30
11 Sales High 0.25 0.30 0.40 0.29866 2,000
12 Sales Medium 0.35 0.40 0.50 0.39866 1,200
13 Sales Low 0.40 0.30 0.10 0.3026801 800
14
16 Sales High 0.35 0.40 0.50 0.39866 1,500
17 Sales Medium 0.40 0.50 0.45 0.4943886 900
18 Sales Low 0.25 0.10 0.05 0.1069514 600
19
21
23 ($thousands)

Chapter 9
DECISION ANALYSIS
The optimal decision if the marketing research company predicts weak competition is
to price high ($50), with expected revenue of $1.521 million.
A B C D E F G H I
3 Medium $40
6 Sales Sales High 0.20 0.25 0.30 0.2796296 2,500
8 High 50 Sales Low 0.55 0.45 0.35 0.3907407 1,000
9 Medium 30
11 Sales High 0.25 0.30 0.40 0.362963 2,000
12 Sales Medium 0.35 0.40 0.50 0.462963 1,200
13 Sales Low 0.40 0.30 0.10 0.1740741 800
14
16 Sales High 0.35 0.40 0.50 0.462963 1,500
17 Sales Medium 0.40 0.50 0.45 0.4592593 900
18 Sales Low 0.25 0.10 0.05 0.0777778 600
19
21
23 ($thousands)
Then, incorporating the expected payoff with each possible prediction by the marketing
company, along with the expected revenue without information from part a, we
combine the whole problem into the following decision tree.
Proceed with no info

1515
1515 1515
0.268
Predict Severe
1 1456.42
1515 1466.41791 1456.41791
0.597
Employ Marketing Research Predict Moderate
1511.15
-10 1505 1521.1474 1511.147404
0.135
Predict Weak
1574.26
1584.25926 1574.259259
Charlotte should not purchase the services of the market research company. The
information is not worth anything since it does not affect the decision. Regardless of
the prediction, the optimal policy is to set the price at $50.

Chapter 9
DECISION ANALYSIS
9.4 a) The available data are summarized in the following spreadsheet.
A B C D
1 Costs Probability
2 ($million) of Success
3 Research 0.3 0.8
4 Development 0.8 0.65
5 Marketing 0.2
6
7 Revenues from R&D
8 ($million)
9 Sell Product Rights 1
10 Sell Research Results 0.2
11 Sell Current DSS 2
12
13 Sales
14 Revenue
15 ($million) Probability
16 High 8 0.3
17 Medium 4 0.5
18 Low 2.2 0.2
b) The basic decision tree is shown below.
No Success (Sell Current DSS)
Do Research Don't Develop (Sell Current DSS & Research Results)
Success No Success (Sell Current DSS & Research Results)
Develop Don't Market (Sell Current DSS & Rights)
Success High
Market Medium
Low
Don't do Research (Sell Current DSS)

Chapter 9
DECISION ANALYSIS
c) The decision tree displays all the expected payoffs and probabilities.
0.2
1.7
2 1.7

1.9
-0.3 2.489 2.2 1.9
0.8 0.35
2 1.1
0 2.69 2.2 1.1
Develop Don't Market (Sell Current DSS & Product Rights)

1.9
-0.8 2.69 3 1.9
0.65 0.3
Success High
1 2 6.7
2.489 0 3.54 8 6.7
0.5
Market Medium
2.7
-0.2 3.54 4 2.7
0.2
Low
0.9
2.2 0.9

2
2 2
d) The best course of action is to do the research project. The expected payoff equals
$2.489 million.

Chapter 9
DECISION ANALYSIS
e) The decision tree with perfect information on research is displayed. The expected value
in this case equals $2.549 million. The difference between the expected values with and
without information equals $60,000, which is the value of perfect information on
research.
No Research (Sell Current DSS)
2
2 2
0.8
Research will be Success Don't Develop (Sell Current DSS & Research Results)
2 1.9
0 2.686 2.2 1.9
0.35
Do Research No Success (Sell Current DSS & Research Results)
2 1.1
-0.3 2.686 2.2 1.1

1.9
-0.8 2.686 3 1.9
0.65 0.3
Success High
2 6.7
2.5488 0 3.54 8 6.7
0.5
Market Medium
2.7
-0.2 3.54 4 2.7
0.2
Low
0.9
2.2 0.9
0.2
Research won't be Successful (Sell Current DSS)
2
2 2

Chapter 9
DECISION ANALYSIS
f) The decision tree with perfect information on development is displayed. The expected
value in this case equals $2.762 million. The difference between the expected values
with and without information equals $273,000, which is the value of perfect
information on development.
2
2 2
0.65 0.2
Development Successful No Success (Sell Current DSS)
2 1.7
0 3.172 2 1.7

1.9
-0.3 3.172 2.2 1.9
0.8
Success Don't Market (Sell Current DSS & Product Rights)
2 1.9
0 3.54 3 1.9
0.3
Develop High
2 6.7
2.7618 -0.8 3.54 8 6.7
0.5
Market Medium
2.7
-0.2 3.54 4 2.7
0.2
Low
0.9
2.2 0.9
0.35
Development would not be Successful (Sell Current DSS)
2
2 2

Chapter 9
DECISION ANALYSIS
g-i) The decision tree with expected utilities is displayed. The expected utilities are
calculated in the following way: for each of the outcome branches of the decision tree
(e.g. profit of $1.7 million) the corresponding utility is calculated (e.g. 9.4737). Once
this is done, the expected utilities are automatically calculated by TreePlan. The best
course of action is not to do the research (expected utility of 10.145 vs. 9.846 in the
case of “Do research”).
0.2
9.4737 Utility
2 9.47 1.7 $million

9.9394 Utility
-0.3 9.846 2.2 9.94 1.9 $million
0.8 0.35
1 7.506 Utility
0 9.94 2.2 7.51 1.1 $million

9.9394 Utility
-0.8 9.55 3 9.94 1.9 $million
0.65 0.3
Success High
2 2 12.46 Utility
10.14 0 10.7 8 12.46 6.7 $million
0.5
Market Medium
11.189 Utility
-0.2 10.7 4 11.189 2.7 $million
0.2
Low
6.5991 Utility
2.2 6.5991 0.9 $million

10.145 Utility
2 10.145 2 $million

Chapter 9
DECISION ANALYSIS
j) The expected utility of doing research, even if we know it will be successful (with
perfect information) equals 9.9394 which is still less than the expected utility of the
alternative “No research” (10.145). Therefore, the best course of action is not to do the
research no matter what, implying a value of zero for perfect information on the
outcome of the research effort.

10.145 Utility
2 10.145 2 $million
0.8
Research will be Success Don't Develop (Sell Current DSS & Research Results)
1 9.9394 Utility
0 10.14469 2.2 9.9394 1.9 $million
0.35
Do Research No Success (Sell Current DSS & Research Results)
1 7.506 Utility
-0.3 9.9394 2.2 7.506 1.1 $million

9.9394 Utility
-0.8 9.55102 3 9.9394 1.9 $million
0.65 0.3
Success High
2 12.46 Utility
10.145 0 10.65 8 12.46 6.7 $million
0.5
Market Medium
11.189 Utility
-0.2 10.652 4 11.19 2.7 $million
0.2
Low
6.5991 Utility
2.2 6.599 0.9 $million
0.2
Research won't be Successful (Sell Current DSS)
10.145 Utility
2 10.14469 2 $million

Chapter 9
DECISION ANALYSIS
k) The expected utility for perfect information on development equals 10.321 which is
more than the expected utility for the case with no information (10.145). The value of
perfect information on development is calculated as the difference between the inverses
of these two utility values (U-1[10.321] - U-1[10.145] = 20.933 – 20 = 0.933. The value
of perfect information on the outcome of the development effort equals $93,300.
10.1447 Utility
2 10.145 2 $million
0.65 0.2
Development Successful No Success (Sell Current DSS)
2 9.47375 Utility
0 10.4164711 2 9.4737 1.7 $million

9.9394 Utility
-0.3 10.416 2.2 9.9394 1.9 $million
0.8
Success Don't Market (Sell Current DSS & Product Rights)
2 9.9394 Utility
0 10.652 3 9.9394 1.9 $million
0.3
Develop High
2 12.4599 Utility
10.321 -0.8 10.652 8 12.5 6.7 $million
0.5
Market Medium
11.1887 Utility
-0.2 10.652 4 11.2 2.7 $million
0.2
Low
6.59908 Utility
2.2 6.6 0.9 $million
0.35
Development would not be Successful (Sell Current DSS)
10.1447 Utility
2 10.1446871 2 $million

Decision Analysis Ba 182

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Decision Analysis Ba 182

Uploaded by

Copyright:

Available Formats

Chapter 9

Jean should buy 12 cases. The maximum expected payoff is $53.60.

f) (i) 0.2 and 0.5

Jean should purchase 12 cases. The maximum expected payoff is $55.20.

Jean should purchase 12 cases. The maximum expected payoff is $54.40.

Jean should purchase 11 cases. The maximum expected payoff is $53.40.

9.4 a) Max(Conservative) = $30 million

Counter-cyclical and conservative cross at approximately p=0.62.

f) When the prior probability of S1 is 0.2, alternative A2 should be chosen, with an

When the prior probability of S1 is 0.6, alternative A1 should be chosen, with an

9.7 a) Max(A1) = $220 thousand, Max(A2) = $200 thousand.

b) Crop 1 has the highest expected payoff of $31,500.

Barbara Miller should pay a maximum of $1,800 to increase x to 75.

9.10 a) Alternative A2 has the highest expected payoff of $1,000.

EP(with perfect information) = (0.2)(4) + (0.5)(2) + (0.3)(1) = $2,100

EVPI = EP(with perfect information) – EP (without more information)

EVPI = EP(with perfect information) – EP (without more information)

9.11 a) Alternative A1 has the highest expected payoff of $35.

EP(with perfect information) = (0.5)($50) + (0.3)($100) + (0.2)(–$10) = $53

EVPI = EP(with perfect information) – EP (without more information)

EVPI = EP(with perfect information) – EP (without more information)

9.12 a) Alternative A3 has the highest expected payoff of $35,000.

EP(with information) = (0.2)(10) + (0.8)(66.25) = 55

EP(with information) = (0.3)(20) + (0.7)(42.857) = 38

d) If S3 occurs for certain then choose alternative A1 (payoff is $100,000).

EP(with information) = (0.5)(100) + (0.5)(10) = 55

EP(with perfect information) = (0.2)(10) + (0.3)(20) + (0.5)(100) = $58,000

EVPI = EP(with perfect information) – EP (without more information)

Dry 0.3 0.6

-100 -50 0.9

-100 190 0.6

-100 100 0.75

Sensit - Sensitivity Analysis - Plot

Sensit - Sensitivity Analysis - Tornado

90 100 110 120 130 140 150 160 170

b) Extending credit maximizes the expected payoff ($8,000).

EP(with perfect information) = (0.2)(0) + (0.5)(10) + (0.3)(20) = $11,000

EVPI = EP(with perfect information) – EP (without more information)

d) PF = Poor Finding AF = Average Finding GF = Good Finding

Prior Conditional Joint Posterior

Don't Extend Credit

Don't Extend Credit

Don't Extend Credit

Don't Extend Credit

9.16 a) Alternative A1 maximizes the expected payoff ($100).

This indicates that it might be worthwhile to do the research.

g) If S1 is predicted, then choosing alternative A1 maximizes the expected payoff

i) Expected payoff given research is (0.36)($233.33) + (0.64)($75) – $100 = $32.

-100 32.06 0.25

b) Choosing to develop the product maximizes the expected payoff ($400,000).

EP(perfect information) = (0.667)(1.5) + (0.333)(0) = $1 million.

EVPI = EP(with perfect information) – EP(without more information)

Sensit - Sensitivity Analysis - Tornado

Profit if Success1.125 1.875

EP(with perfect information) = (0.8)(–$750) + (0.2)(–$1,500) = –$900

EVPI = EP(with perfect information) – EP(without more information)

e) The optimal policy is not to pre-screen or screen.

-125 -1392.95 0.835

f) EVSI = EP(with information ignoring cost of information) – EP(without information)

c) They should build computers, with an expected payoff of $27 million.

0.2 0.4 0.6 0.8 1.0

Build and Sell cross when –54p + 54 = 15 or 54p = 39 or p = 0.722