6. Decision Theory

1. Payoff Matrix: a table of payoffs to analyze the value of each decision we make

2. Various decision making techniques: Maximax, Maximin, Laplace, Maximize Expected Monetary Value (EMV), Minimax Regret

Decision Criteria:

• Maximax (suitable for aggressive DM)

Choose the alternative with the best possible payoff

• Maximin (suitable for conservative DM, Murphy’s law)

Choose the alternative with the best of the worst possible payoffs

• Laplace (suitable for agnostic DM) 

Choose the alternative with the best average payoff assuming all outcomes are equally likely

• Minimax regret (Lawyer’s rule)

Choose the alternative that has the least of the worst regrets

• Max EMV (correct choice for a risk-neutral DM!)

Determine the expected payoff of each alternative, and choose the alternative that has the best expected payoff

• Decision Tree: a visual aide to help you understand sequential decision makings

Value of Perfect Information: your willingness to pay for the perfect information if it is available

3. Expected value of perfect information (EVPI)

The difference between the expected payoff with perfect information and the expected payoff under risk

• EVPI = expected value (payoff) with perfect information – expected value (payoff) under uncertainty