Halevy, Yoram: A Bayesian Approach to Uncertainty Aversion
World Conference Econometric Society, 2000, Seattle

Vincent Feltkamp, Statistics Netherlands
Yoram Halevy, University of British Columbia
A Bayesian Approach to Uncertainty Aversion
Session: C-9-25  Monday 14 August 2000  by Halevy, Yoram
The Ellsberg paradox demonstrates that people's belief over uncertain events might not be representable by subjective probability. We argue that Uncertainty Aversion may be viewed as a case of "Rule Rationality''. This paradigm claims that people's decision making has evolved to simple rules that perform well in most regular environments. Such an environment consists of replicas of some basic singular circumstance. When the rule is applied to a singular environment, the behavior may seem paradoxical. We claim that the regular environment in which decisions under uncertainty take place, is described by one decision that spans multiple ambiguous risks, which are positively correlated. We show that when a risk averse individual has a Bayesian prior and uses a rule, which is optimal for the regular ambiguous environment, to evaluate a singular vague circumstance - his behavior will exhibit uncertainty aversion. Thus, the behavior predicted by Ellsberg may be explained within the Bayesian expected utility paradigm.
Submitted paper full-text in .pdf


File created by Jurgen Doornik with eswc2000.ox on 2-01-2001