A Mixture Model of Willingness to Pay Distributions

In this paper we propose a mixture model of willingness to pay distributions for contingent valuation studies. By allowing a point mass at zero, this model nests the conventional model as a special case. We discuss both parametric and non- parametric estimations of the mixture model. We consider estimation under two different data information settings for a double bounded dichotomous choice format. The implications of the mixture model in the estimation of the mean and the median of the distribution are presented.