The problem with decisions is that we rarely, if ever, find out if our decisions were good or bad. Was choosing your major, for instance, a good decision or could you have made a better one? I don’t think most of us would ever know the answer to this question. So, is it possible that we regularly make bad decisions but don’t know that we do? And, if so, how can we fix something if don’t know it is broken?
In fact, we do regularly make bad decisions. This has been shown in many experimental studies some of which will be covered in this class. What is more, for some types of decision problems we are hardwired to make mistakes. This means that we are bound to go wrong regardless of how much we know or how smart we are. So, what can we do to remedy this problem? Quite a bit, as it turns out.
What follows are a few examples of some decision problems and more general topics that will be covered.
EXAMPLE 1: One rule of rational decision making is that our choice should not be affected by irrelevant factors. If we prefer A over B we should choose A over B no matter what is the context they are presented in. Do you think our choices satisfy this condition? And if they don’t—which, actually, is the case—can you see how it can be used to manipulate what you buy in a supermarket or what you vote for in a referendum?
EXAMPLE 2: Suppose you are asked to choose one of the following two options (A) get 1m (1 million dollars) for sure and (B) get 1m with probability 0.89, 2.5m with probability 0.1 and 0 with probability 0.01. And suppose that you are also asked to choose one from a pair of two other options: (C) get 1m with probability 0.11 and 0 with probability 0.89 and (D) get 2.5m with probability 0.1 and 0 with probability 0.9. Is there anything wrong with choosing A and C? And, what would you say about choosing A and D? Similarly, would you find anything wrong with choosing B and C or B and D? In an experimental study Maurice Allais (1988 Nobel Prize in economics) has asked people to make such choices. The outcome of this experiment—it became one of the most conspicuous studies in economics—is known as Allais Paradox. Can you guess what the nature of the finding is and why it is so important?
EXAMPLE 3: This is more of a question than an example but the topic may be of interest to some of you perhaps: There is one decision procedure I have seen used by all graduate admission committees. I won’t tell you what the procedure is but I will tell you that it was famously described and studied by Herbert Simon (1978 Nobel Prize in economics.) If you know this procedure you can devise simple but effective ways to increase your chances of admission.
READINGS: There is one textbook that will be regularly used: Itzhak Gilboa’s Making Better Decisions: Decision Theory in Practice, Wiley, 2011.
I will post online excerpts from other sources which include Reid Hastie and Robyn Dawes, Rational Choice in an Uncertain World: The Psychology of Judgment and Decision Making, Sage, 2010; Avinash Dixit and Barry Nalebuff The Art of Strategy, Norton, 2008; Kenneth Williams, Game Theory a Behavioral Approach, Oxford University Press, 2013, David Kreps, Notes on the Theory of Choice, Westview Press, 1988, Raymond Wilder “The Axiomatic Method,” pages 1621-1640 in The World of Mathematics, Simon and Schuster, 1956; Analyzing Politics by Kenneth Shepsle, Norton, 2010; Avinash Dixit, Susan Skeath and David Reiley, Games of Strategy, Norton, 2015.
GRADING: Three short tests (50%), three homework assignments (25%) and class participation (25%.)
CLASS FORMAT: A good part of the class time will be spend in groups in which you will be solving problems or working on short projects. (Group membership will change every week and will be determined by a random mechanism.) Members of successful groups will get extra credit points. Points will be accumulated over the semester and used in grading your class participation.
An optional final exam is the only way to improve your grade.
FINAL EXAM: An optional way to improve your grade is by taking the final exam. Final exam will count for 50% of your test grade. For example, suppose your average test score is 86%. If you decide not to take the final exam, your class grade will be calculated with the 86% test average counting as 50% of your class grade. If, however, you take the final exam and score 94% on it, your class grade will be calculated with 0.5*86% + 0.5*94% = 90% counting as 50% of your class grade.
CRIB SHEET: All testing is closed book but you ARE ALLOWED to have a crib sheet—a single standard size sheet of paper with whatever information you want to put on it (both sides.)