Decision Theory

Module 4
Objectives

• Compute problems involving mathematical expectation/ expected

value
• List the characteristics of a decision theory approach to decision
making
• Describe and give examples of decision under certainty, risk and
complete uncertainty
• Determine the expected value of perfect information
Decision Theory

• Is a methodology for making management decisions which

constitutes a particular branch of decision science

• Analysis of decision situations in which certainty cannot be

assumed

• Decision theory aids the decision maker address the problem of

making complex decisions under uncertain conditions
Steps in Decision Making

• 1. Clearly define the problem

• 2. List all the possible alternative events that may occur
• 3. Determine all possible outcome
• 4. Identify the profit/loss of each combination of alternative
events and outcomes
• 5. Choose one mathematical decision theory model
• 6. Apply the model and make the decision
Decision Making Criteria

• Maximax Criterion/Optimistic Approach

• Maximizes the maximum payoff for the different decisions starting with the
identification of maximum payoffs of each alternative decision. The
maximax decision is that decision which yields the largest maximum payoff
• Maximin Criterion/ Conservative Approach
• Simply maximizes the minimum payoffs given to various decisions that are
possible. It is a two-step process once the payoff table has been
formulated. The first step is to identify the minimum payoff for each
decision. The second step is to select the largest minimum payoff
• Minimax Regret Criterion
• Involves the construction of an opportunity loss or regret matrix prior to
applying the minimax rule. Construct an opportunity loss table by
transforming each element in the payoff table to an opportunity loss. The
degree of an opportunity loss for a given element is the loss incurred by not
selecting the optimal alternative decision given a state of nature.
• Steps
• 1. Find the largest element in the first column
• 2. Subtract each element in the column from the largest element to compute
the opportunity loss
• 3. Choose the decision with the smallest maximum regret
Example
• A farmer in Region 2 must decide which crop to plant next year on his land; corn, peanuts or soy
beans. The return of each crop will be determined by whether a new trade bill with Hongkong
passes the Senate. The profit the farmer will realize from each crop given the two possible results
on the trade bill is shown in the following payoff table.

Trade Bill
Crop Pass Fail
Corn 2,000,000 700,000
Peanuts 1,100,000 500,000
Soybeans 1,400,000 1,200,000

• Determine the best crop to plant using the following decision criteria
• A. Maximax B. Maximin C. Minimax Regret
Solution

A. Maximax Criteria • Step 2. Identify the highest maximum payoff

Step 1. Select the column with
maximum payoff
<=maximum

• Decision: The planting of the corn will result in a

Maximum Payoff maximax payoff of 2,000,000
• B. Maximin Criterion • Step 2. Identify the largest
• Step 1. Select the column with minimum payoff
the minimum payoff

• Decision: The planting of

Minimum payoff soybeans will result to a
minimum payoff of 1,200,000
 Trade Bill
Crop Pass Fail

C. Minimax Regret Criterion

Corn 0 500,000
Step 1. Determine the largest element in first column
Peanuts 900,000 700,000
Soybeans 600,000 0

Step 3. Identify the maximum regret

Step 2. Subtract every element from the largest payoff to both
columns
Pass
Corn =2,000,000-2,000,000=0 Crop Maximum Regret
Peanuts =2,000,000-1,100,000=900,000
Soybeans =2,000,000-1,400,000=600,000 Corn 500,000 =minimum
Fail
Corn =1,200,000-700,000=500,000
Peanuts =1,200,000-500,000=700,000 Peanuts 900,000
Soybeans =1,200,000-1,200,000=0
Soybeans 600,000

Decision: The planting of corn will result in at most 500,000 in

regret
Expected Value/ Mathematical Expectation

• Of a discrete random variable is very important in characterizing

its probability distribution.
• Originally the concept of expected value was introduced with
reference to game of chance where, if a player stands to win an
amount k with probability P, his expected value or expectation is
defined as k(P)
• Example
• If a man purchases a raffle ticket, he will win a first prize of P500,000 or a
second prize of P200,000 with probabilities 0.0001 and 0.0005. What should
be a fair prize to pay the ticket?
Solution

• EV=(500,000)(0.0001)+(200,000)(0.0005)
• =50+100
• =P150
• The fair price for the ticket is P150
• Example
• A gambler tosses a die. If 1,3 or 5 appears, the gambler will be paid P9, if a 2 or 4
turns up, he will lose P9, and if 6 is showing, he will win P12. Determine the
expectation of the gambler.
• Solution
• P(A)=probability of 1,3 and 5 will appear……. or 3/6
• P(B)= probability of 2 or 4 will appear………….or 2/6
• P(C)=probability of 6 will appear………………….or 1/6
• EV=9P(A)-9P(B)+12P(C)
• =9(3/6)-9(2/6)+12(1/6)
• =4.5-3+2
• =P3.5
• The gambler expects to win P3.5 per toss of a die
Decision Trees

• Another useful technique in analyzing a decision situation

• It is a graphical diagram consisting of nodes and branches
• The user computes the expected value of each outcome and makes a decision
• The main benefit of a decision tree is that it provides a graphic of the decision
making process
• The circles and squares are referred as nodes while the branches reflect the
alternative decisions possible
• Decision Variable
• Is a variable whose value represents a potential decision on the part of the decision maker
Example
• The Omega Corp, a corporate raider, has acquired a shoe company and is contemplating the
future of one of its major plants located in Marikina. Three alternative decisions are being
considered. (1) expand the plant and produce lightweight, durable shoes for possible sale at a
department store, a market with a little foreign competition;(2) maintain the status quo at
the plant; continuing production of shoes that are subject to heavy foreign competition; or
(3) sell the plant now. If one of the first two alternatives is chosen, the plant will still be sold
at the end of the year. The amount of profit that could be earned by selling the plant in a
year depends on foreign market conditions, including the status of a trade embargo bill in
Congress. The following payoff table describes the situation.

States of Nature
Decision Good Foreign Bad Foreign Competitive
Competitive Condition Condition
Expand 30,000,000 20,000,000
Maintain status quo 50,000,000 -7,500,000
Sell now 15,000,000 15,000,000
• A. Assume that it is now possible to estimate a probability of 0.75
that good foreign competitive conditions will exist and a
probability of 0.25 that the bad conditions will exist. Determine
the best decision using expected value and expected opportunity
loss.
• B. Compute for the Expected Value of the Perfect Information
• C. Develop a decision tree for this situation, with expected value
at the probability nodes
• A.1 Decision with expected value and Expected Opportunity Loss
• Step 1. Determine the Expected Value by Substituting the estimate probability
• Expand= 30,000,000(0.75)+20,000,000(0.25)
• =22,500,000+5,000,000
• =27,500,000
• Status Quo =50,000,000(0.75)-7,500,000(0.25)
• =37,500,000-1,875,000
• =35,625,000
• Sell =15,000,000(0.75)+15,000,000(.25)
• = 11,250,000+3,750,000
• =15,000,000
• Step 2. Select the highest solution and make the decision
• Decision: The decision maker should maintain status quo since it generates the highest expected
value
 • Step 2. Substitute Opportunity loss to the estimate probability

• A.2 Expected Opportunity Loss • Expand=20,000,000(0.75)+0(0.25)

• =15,000,000
• Step 1. Determine the • Status Quo= 0(0.75)+27,500,000(0.25)

opportunity loss table • =0+6,875,000

• =6.875,000
States of Nature • Sell =35,000(0.75) +5,000,000 (.25)
Decision Good Foreign Bad Foreign • =26,250,000+ 1,250,000
Competitive Competitive • =27,500,000
Condition Condition
• Step 3. Select the lowest result and make the decision
Expand 20,000,000 0
Maintain 0 27,500,000
Status Quo • Decision: The decision maker
Sell Now 35,000,000 5,000,000 should maintain status quo since
it generates the lowest expected
opportunity loss
B. Expected Value of Perfect Information

• Step 1. Determine the expected value given perfect information

• =50,000,000(0.75)+20,000,000(0.25)
• =37,500,000+5,000,000
• =42,500,000
• Step 2. Compute for the expected value without perfect information
• =50,000,000(0.75)-7,500,000(0.25)
• =37,500,000-1,875,000
• =35,625,000
• Step 3. Solve for the expected value of perfect information
• EVPI=42,500,000-35,625,000
• =6,875,000
• The EVPI, 6,875,000 is the amount that the investor would pay to purchase perfect information from some
other sources, such as economic forecaster
C. Decision Tree

• Step 1. Construct a decision tree

• Step 2. Compute for the expected value of the nodes
• Step 3. Select the highest solution among the circle nodes and
make the decision

The decision is to maintain the status quo, where an expected

payoff of 35,625,000, is the same result in the expected value
criterion.
Practice Exercise 1

• A farmer in Bulacan is considering either leasing some extra land or investing in savings certificates at
local bank. If weather conditions are good next year, the extra land will allow the farmer to have an
excellent harvest. The savings certificate will result in the same return regardless of weather conditions.
The return of each investment given each type of weather condition is shown in the following payoff table

Weather
Good Bad
Lease land 450,000 -200,000
Buy Savings Certificate 50,000 50,000

• Select the best decision using the following decision criteria

• A. Maximax B. Maximin ` C. Minimax Regret
Answer
• A. Maximax
• Determine the best possible payoff
• Decision: Leasing of land will result to a maximax payoff of P450,000
• B. Maximin
• Determine the worst payoff on each alternative, then choose the best among the worst
• Decision: Buying of certificate will result to maximin payoff of P50,000
• C. Minimax Regret
• Opportunity loss table
Lease land 0 250,000 250,000
Buy savings 400,000 0 400,000
certificate
• Decision: Leasing of land in at most, P250,000 in regret
Practice Exercise 2

• A bellboy of the Alpha Hotel in Manila has been offered a management position.
Although accepting the offer would assure him a job if there were recession, if
good economic condition prevailed he would actually make less money as a
supervisor than as a bellboy ( because of the large tips he gets as a bellboy). His
salary during the next five years for each job given each future economic
condition is shown in the following payoff table
Economic Conditions
Decision Good Recession
Bellboy 2,000,000 1,100,000
Supervisor 1,500,000 1,500,000

• Select the best decision using the following decision criteria

• A. maximax B. Maximin C. Minimax Regret
Answer
• A. Maximax
• Determine the best possible payoff
• Decision: As Bellboy will result to a maximax payoff of P2,000,000
• B. Maximin
• Determine the worst payoff on each alternative, then choose the best among the worst
• Decision: As Supervisor will result to maximin payoff of P1,500,000
• C. Minimax Regret
• Opportunity loss table
Bellboy 0 400,000 400,000
Supervisor 500,000 0 500,000

• Decision: As Bellboy will result to P400,000 in regret

Practice Exercise 3

• The owner of the Bella Construction Company must decide among building a housing development, constructing a
shopping center or leasing all the company’s equipment to another company. The profit that will result from each
alternative will be determined by whether material cost remain stable or increase. The profit from each
alternative given two possibilities for material costs is shown in the following payoff table.

Material Costs
Decision Stable Increase
Houses 3,500,000 1,500,000
Shopping Center 5,250,000 1,000,000
Leasing 2,000,000 2,000,000

• Determine the best decision using the following decision criteria

• A. Maximax B. Maximin C. Minimax Regret
Answer

• A. Maximax
• Determine the best possible payoff
• Decision: Shopping Center will result to a maximax payoff of P5,250,000
• B. Maximin
• Determine the worst payoff on each alternative, then choose the best among the worst
• Decision: Leasing will result to maximin payoff of P2,000,000
• C. Minimax Regret
• Opportunity loss table

Houses 1,750,000 500,000 1,750,000

Shopping Center 0 1,000,000 1,000,000
Leasing 1,500,000 0 1,500,000

• Decision: Shopping Center will result to P1,000,000 in regret

Practice Exercise 4 ( 3.22.2021)

• Consider the following payoff table which illustrates capacity planning problem.

Possible Future Demand

Alternative Low Moderate High
Small Facility P10,000,000 P10,000,000 P10,000,000
Medium Facility 7,000,000 12,000,000 12,000,000
Large Facility (4,000,000) 2,000,000 16,000,000
• A. Compute for maximin, maximax, minimax regret
• B. Using these probabilities: low=.30, moderate=.50 and high =.20, determine expected
value and expected opportunity loss
• C. Compute for the EVPI
Answer

• A. maximin= P10,000,000, build the small facility

• maximax=P16,000,000, build the large facility
• minimax regret =P4,000,000, build medium facility
• B1.Expected Value
• Small =.30 (10,000,000)+.50(10,000,000)+.20(10,000,000)=P10,000,000
• Med =.30(7,000,000)+.50(12,000,000)+.20(12,000,000)=P10,500,000
• Large =.30(-4,000,000)+.50(2,000,000)+.20(16,000,000)=P3,000,000
• Decision: Choose medium facility because it has the highest expected value
• B2. Opportunity Loss
• Small =.30(0)+.50(2,000,000)+.20(6,000,000)=P2,200,000
• Med =.30(3,000,000)+.50(0)+.20(4,000,000)=P1,700,000
• Large =.30(14,000,000)+.50(10,000,000)+.20(0)=P9,200,000
• Decision: Should build medium facility since it generates the
lowest expected opportunity loss
• C. Expected value of Perfect Information
• Step 1. Determine expected value with perfect information
• =10,000,000(.30)+12,000,000(.50)+16,000,000(.20)
• =P12,200,000
• Step 2. Computed expected value without perfect information (highest expected monetary value)
• =(7,000,000).30+(12,000,000).50+(12,000,000).20 =P10,500,000
• Step 3. Subtract
• =12,200,000-10,500,000
• =P1,700,000
• P1,700,000 is the maximum amount that the investor would pay to purchase perfect information