The Predictive Utility of Generalized Expected Utility Theories

Abstract

Many alternative theories have been proposed to explain violations of expected utility \n(EU) theory observed in experiments. Several recent studies test some of these alternative \ntheories against each other. Formal tests used to judge the theories usually count the \nnumber of responses consistent with the theory, ignoring systematic variation in responses \nthat are inconsistent. We develop a maximum-likelihood estimation method which uses \nall the information in the data, creates test statistics that can be aggregated across studies, \nand enables one to judge the predictive utility-the fit and parsimony-of utility theories. \nAnalyses of 23 data sets, using several thousand choices, suggest a menu of theories which \nsacrifice the least parsimony for the biggest improvement in fit. The menu is: mixed \nfanning, prospect theory, EU, and expected value. Which theories are best is highly \nsensitive to whether gambles in a pair have the same support (EU fits better) or not (EU \nfits poorly). Our method may have application to other domains in which various theories \npredict different subsets of choices (e.g., refinements of Nash equilibrium in noncooperative \ngames).