354. Decision Analysis

A framework for making decisions under uncertainty. Distinguishes:

Decision analysis asks: given uncertainty, how should I choose?

354.1. Core components

  1. Decision tree — graphical representation of the decision problem
  2. EMV — Expected Monetary Value, the standard decision criterion
  3. EVPI — value of perfectly knowing the state in advance
  4. EVSI — value of imperfect (sample) information
  5. Decision criteria — alternatives to EMV under deep uncertainty
  6. Utility theory — risk-averse decision making

354.2. Steps in a decision analysis

  1. Frame: identify objectives, alternatives, uncertainties, payoffs
  2. Structure: build a decision tree (or influence diagram)
  3. Quantify: assign probabilities and payoff values
  4. Evaluate: compute expected value of each strategy
  5. Sensitivity analysis: how does the answer change with input changes?
  6. Recommend: highest expected value (or by other criterion)

354.3. When to use

354.4. Limits

354.5. See also