ARE PEOPLE BAYESIAN - UNCOVERING BEHAVIORAL STRATEGIES

Economists and psychologists have recently been developing new theorie s of decision making under uncertainty that can accommodate the observ ed violations of standard statistical decision theoretic axioms by exp erimental subjects. We propose a procedure that finds a collection of decision rules that best explain the behavior of experimental subjects . The procedure is a combination of maximum likelihood estimation of t he rules together with an implicit classification of subjects to the v arious rules and a penalty for having too many rules. We apply our pro cedure to data on probabilistic updating by subjects in four different universities. We get remarkably robust results showing that the most important rules used by the subjects (in order of importance) are Baye s's rule, a representativeness rule (ignoring the prior), and, to a le sser extent, conservatism (overweighting the prior).