Economists and psychologists have recently been developing new theories of decision making under uncertainty that can accommodate the observed violations of standard statistical decision theoretic axioms by experimental subjects. We propose a procedure which finds a collection of decision rules that best explain the behavior of experimental subjects. The procedure is a combination of maximum likelihood estimation of the rules together with an implicit classification of subjects to the various rules, and a penalty for having too many rules. We prove that our procedure yields consistent estimates (as the number of tasks per subject and the number of subjects go to infinity) of the number of rules being used, the rules themselves, and the proportion of our population using each of the rules. We apply our procedure to data on probabilistic updating by subjects in four different universities. We get remarkably robust results which show that the most important rules used by the subjects are Bayes rule, representativeness rule (ignoring the prior), and conservatism (over-weighting the prior). In our procedure, the subjects are allowed to make errors, and our estimated error rate is typically 20%.
Stored only when both papers are in the corpus — 8 citations exist upstream.