ESTIMATION OF REVEALED PROBABILITIES AND UTILITY-FUNCTIONS FOR PRODUCT-SAFETY DECISIONS

Using survey data on consumer product purchases, this paper introduces an approach to estimate jointly individual utility functions and risk perceptions implied by their decisions. The behavioral risk beliefs r eflected in consumers' risky decisions differ from the stated probabil ities given to them in the survey, These results are not consistent wi th a Bayesian learning model in which the information respondents util ize is restricted to what the survey presents. The results are, howeve r, potentially consistent with models in which prior risk information is influential or models in which people do not act in a fully rationa l manner.