Decision Making Theory
Decision Making Theory
Learning Objectives
List steps of decision making process. Describe different types of decision making environments. Make decisions under uncertainty when probabilities are not known. Make decisions under risk when probabilities are known. Use Excel for problems involving decision tables. Develop accurate and useful decision trees. Use TreePlan to set up and analyze decision tree problems with Excel. Revise probability estimates using Bayesian analysis. Understand the importance and use of utility theory in decision making.
8.1 Introduction
Decision theory is analytic and systematic approach to study of decision-making. What makes difference between good and bad decisions? Good decisions may be defined as:
Based on logic, Considered all possible decision alternatives, Examined all available information about future, and Applied decision modeling approach.
Type 3: Decision Making under Risk. Decision maker has some knowledge regarding probability of occurrence of each outcome or state of nature. Examples:
Probability of being dealt club from deck of cards is 1/4. Probability of rolling 5 on die is 1/6.
2. Maximin.
3. Equally likely. 4. Criterion of realism. 5. Minimax regret. First four criteria calculated directly from decision payoff table. Fifth minimax regret criterion requires use of opportunity loss table.
Maximax Criterion
Maximax criterion selects alternative maximizes maximum payoff over all alternatives. First locate maximum payoff for each alternative. Select alternative with maximum number. Decision criterion locates alternative with highest possible gain. Called optimistic criterion. Table shows maximax choice is first alternative: "construct large plant." $200,000 payoff is maximum of maximum payoffs for each decision alternative.
Thompson Lumber Company Maximax criterion selects alternative that maximizes maximum payoff over all alternatives. First alternative, "construct a large plant, $200,000 payoff is maximum of maximum payoffs for each decision alternative.
Maximax Criterion
Maximin Criterion
Maximin criterion finds alternative maximizes minimum payoff over all alternatives. First locate minimum payoff for each alternative. Select alternative with maximum number. Decision criterion locates alternative that has least possible loss. Called pessimistic criterion. Maximin choice, "do nothing," is shown in table. $0 payoff is maximum of minimum payoffs for each alternative.
Maximin Criterion
Thompson Lumber Company Maximin criterion finds alternative maximizes minimum payoff over all alternatives. First locate minimum payoff for each alternative, and select alternative with maximum number.
Equally likely, also called Laplace, criterion finds decision alternative with highest average payoff. Calculate average payoff for every alternative. Pick alternative with maximum average payoff. Assumes all probabilities of occurrence for states of nature are equal. Equally likely choice is second alternative, "construct a small plant." Strategy shown in table has maximum average payoff ($40,000) over all alternatives.
Criterion of Realism
Formula for criterion of realism = a x (maximum payoff for alternative) + (1-a) x (minimum payoff for alternative) Assume coefficient of realism a = 0.80. Best decision would be to construct a large plant.
Criterion of Realism
Thompson Lumber Company Coefficient of realism a = 0.80. $124,000 = (0.80)($200,000) + (0.20)(- $180,000).
EOL Decision
Thompson Lumber Company EOL values are computed as shown. Using minimum EOL as decision criterion, best decision would be second alternative, "construct a small plant" with an EOL of $60,000. Minimum EOL will always result in same decision alternative as maximum EMV.
Using Excel to Solve Decision Making Problems Under Risk Formulas View
Using Excel to Solve Decision Making Problems Under Risk Formulas View
Decision Tree
Thompson Lumber Company Tree usually begins with decision node. Decision is determine whether to construct large plant, small plant, or no plant. Once decision is made, one of two possible states of nature (favorable or unfavorable market) will occur.
problem.
Numbers (e.g., 0.78, 0.22, 0.27, 0.73 ) can be assessed by manager based on experience and intuition. They can be derived from historical data or computed from other available data using Bayes theorem.
Probabilities are based only on cases in which decision to build a plant or take some course of action is actually made.
Conditional probability information is not quite as accurate as desired.
Summary
Introduced decision theory to study decision making. Studied (1) decision making under certainty, (2) decision making under uncertainty, and (3) decision making under risk. Identified best alternatives using criteria: maximax, maximin, equally likely, criterion of realism, and minimax regret. Discussed computation and use: expected monetary value (EMV), expected opportunity loss (EOL), and expected value of perfect information (EVPI). Decision trees were used for larger decision problems in which decisions had to be made in sequence. Computed expected value of sample information (EVSI). Bayesian analysis used to revise or update probability values. Discussed how decision trees can be set up and solved using TreePlan, an Excel add-in.