Course Objectives and Content:

The aim of this course is to introduce and develop the Ramsey-Savage expected utility theorem as a general framework for decision-making under uncertainty. We will discuss utility theory, subjective probability and its elicitation, basic statistical decision theory, sequential decision problems, expert opinions and behavioral paradoxes and their possible resolution. Some applications in the economics of organisation and uncertainty will also be discussed.

The course will emphasise the theoretical foundations of rational choice under uncertainty.

Audience:

Students from business administration, economics and related areas, psychology, statistics, mathematics and engineering with an interest in behavior under uncertainty will find this course useful. The prerequisites are a working knowledge of calculus and prior exposure to probability and statistical inference.

Texts:

Required:

1.David Kreps, Notes on the Theory of Choice, Westview Press, Boulder, Colorado, 1988.

2.Detlof von Winterfeldt and Ward Edwards, Decision Analysis and Behavioral Research, Cambridge University Press, New York, 1986. (Both of these are available in paperback.)

3. Course packet: This contains Ch.2, 4 and 8 of Howard Raiffa's classic Decision Analysis, a chapter on risk aversion from Keeney and Raiffa's Decisions with Multiple Objectives (Cambridge University Press, 1995), Pratt, Raiffa and Schlaifer's "Foundations of Decision Making under Uncertainty", and other relevant papers and book chapters that will be referred to in class. While I have tried to make this as complete as possible, not all material to be covered in class could be included. The use of the library will be necessary on a regular basis.

The two texts adopt different approaches to the subject. Kreps focusses on the rationale for expected utility theory and presents a rigorous mathematical exposition. Von Winterfeldt and Edwards discuss extensions of the basic framework, some inferential problems and behavioural decision theory as well. You may find one approach more to your liking, but both are valid approaches.

Optional:

H. Raiffa (1968): Decision Analysis, Random House/McGraw Hill

A classic introduction to the field. We shall use Ch. 1-4 and 8 in class. The rest of the book is useful for those who find Kreps hard going.

Grading:

The grading will be based on two examinations, each worth 30%, two required homework sets, each worth 10% and a short term paper worth 20%. Class participation and subjective evaluation will be used at the margin in determining grades.

Topics to be covered.

1. Introduction to decision analysis; expected value, decision trees, Bayes' Theorem, expected value of perfect and imperfect information. (Raiffa: Ch.1-3, in course packet. Also Kreps Ch.1, vWE Ch. 1-3.)

2.The von Neumann-Morgenstern Expected Utility Theorem. (Kreps Ch.5, Pratt et al. in course packet.)

3 Utility functions for money, risk aversion and insurance.(Pratt "Risk Aversion in the Small and in the Large"Econometrica 1964, Kreps Ch.6, Keeny and Raiffa in course packet, other readings.)

4. Subjective probability, the Ramsey-Savage Expected Utility Theorem. (Pratt et al. in course packet, Kreps, Ch. 4, 7, 8, 9, vWE Ch.4.)

5. Analysis of decision trees. An introduction to the normal form.

6.Bayes' Theorem and conditional probabilities- foundations. (vWE Ch.5,6. Kreps, Ch. 10, 11.)

7. Expert opinions, " Reaching a Consensus" (De Groot, JASA, 1974 in course packet, Raiffa paper in course packet, Raiffa Ch. 8.)

7. Comparison of information structures. Blackwell's theorem. Information and communication in organizations and team theory.

8. Alternative approaches to uncertainty- belief functions and probabilistic logic. (If time permits.)

9. Bayesian inference, an introduction. . (vWE Ch. 5, 6 .)

10. Preposterior analysis, sequential analysis.

11. Sequential decision-making and search theory. (Ross : Introduction to Stochastic Dynamic Programming Ch. 3 in course packet, Campbell and Lindner paper, Bhattacharya, Chatterjee and Samuelson paper, both in course packet.)

12. Other topics in utility theory, if time permits.

Whenever no readings are listed, this is an indication that library search may be required.

Note: I expect to have to be out of town on October 29 and one other day in the term. Class and office hours on these days will be cancelled, and class will be rescheduled at a convenient time.

The Smeal College of Business Administration welcomes persons with disabilities to all of its classes, programs, and events. If you need accommodations or have questions about access to buildings in which Smeal College activities are held, please contact us in advance of your participation or visit. If you need assistance during a class, program, or event, please contact the member of our staff or faculty in charge.

This publication is available in alternative media on request.

 GO BACK TO DECISION ANALYSIS COURSES AND SYLLABI