Discrete choice under preference uncertainty: an improved structural model for contingent valuation.

Abstract This paper is concerned with discrete choice contingent value estimation in the presence of response errors. We attempt to extend the theoretical frameworks developed by Hanemann (1984, Amer. J. Agric. Econom. 66 , 332-341) and Cameron (1988, J. Environ. Econom. Management 15 , 355-379) by introducing preference uncertainty with individual respondents. We assume that respondents have incomplete knowledge about their true valuation of a nonmarket resource and thus may give wrong yes/no answers on whether or not to pay a given bid amount for the resource. A post-decisional confidence measure for each observed yes/no answer is elicited for characterizing the degree of preference uncertainty. With the help of this confidence measure, we develop an improved structural model for deriving value estimates. Data from a forest environment valuation survey are used for the empirical estimations.