Parametric and Semi-Nonparametric Estimation of Willingness-To-Pay in a Contingent Valuation Framework

Standard procedures for extracting willingness-to-pay (WTP) from dichotomous choice CV questionnaires rely heavily upon parametric assumptions regarding the distribution and form of WTP in the sampled population. However, theory provides little guidance regarding which parametric specification to use and the resulting WTP estimates can be sensitive to the selections made. Here we compare and contrast several parametric and semi-nonparametric estimators that have been proposed in the literature, examining the sensitivity of the resulting WTP estimates to the underlying distribution of preferences and the estimation procedure employed.