Professor Gordon B. Hazen
IE/MS Department, MLSF 3083
Phone 491-5673
Email: Hazen@iems.nwu.edu
An introduction to the prescriptive and descriptive analysis of decision
making under uncertainty. Topics include
Fundamentals: Problem formulation; Decision Trees; Influence
diagrams; Bayes' rule; The value of information. Uncertainty and
its measurement: Subjectivist versus frequentist views of probability;
Axioms for probability; Probability elicitation; Applications: A suicide
prevention model, Assessing the risk of a landslide. Bayesian approaches:
Bayes' theorem; Prior probabilities; The principle of stable estimation;
The likelihood principle; Conjugate distributions; Predictive distributions;
Hypothesis testing; Applications: Predictions of oil spills, Hurricane
seeding; Uncertainty about probability. Single-attribute utility
and value theory: Axioms for preference; Certainty equivalents
and risk aversion; The coefficient of risk aversion; Exponential utility
and the delta property; Criticisms of utility theory: the Allais and Ellsberg
paradoxes. Multiattribute utility theory: Preferential independence
and additively separable value functions; Marginality and additively separable
utility functions; Utility independence and multiplicatively separable
utility functions. Decision analysis applications: Evaluation
of pumped storage sites; Setting pollution standards; A bypass surgery
decision; A capital investment decision.
Text:
Detlof von Winterfeldt and Ward Edwards, Decision Analysis and Behavioral Research, Cambridge University Press 1986.
Course grade based on: Homework exercises, midterm exam, final exam.
Prerequisites: Calculus-based probability including some familiarity with the Poisson process (IE D60-1 sufficient but not required); some familiarity with mathematical proofs.
Instructor:
Professor Gordon B. Hazen
IE/MS Department, MLSF 3083
Phone 491-5673
Email: Hazen@iems.nwu.edu
