ABSTRACT: Making use of restrictions imposed by equilibrium, theoretical progress has been made on the nonparametric and semiparametric estimation and identification of scalar additive hedonic models and scalar nonadditive hedonic models. However, little is known about the practical aspects of estimating such models or of the characteristics of equilibrium in such models. This paper presents computational and analytical results that fill some of these gaps. We simulate and estimate examples of equilibrium in the additive hedonic models and provide evidence on the performance of a maximum likelihood estimation technique. We also simulate examples of equilibria in nonadditive models and provide evidence on the performance of the nonadditive estimation techniques developed by Heckman, Matzkin, and Nesheim (unpublished working paper, 2002).

INTRODUCTION

Hedonic models are general equilibrium frameworks that characterize the pricing of differentiated goods, viewed as bundles of attributes, and the demand and supply of those goods (attributes) under different assumptions about preferences and technology. They allow a systematic economic analysis of the demand and supply of quality. Quality includes enhancement of the attributes of a good embodied in a unit of the good (such as the attributes of a house or a car, as in Rosen, 1974), characteristics of a job (risk or unpleasantness as in Tinbergen, 1956; Sattinger, 1975, 1980, 1993; Thaler and Rosen, 1975) or the amenities offered by an environmental or recreational improvement (as in Smith and Huang, 1995; Banzhaf et al., 2000).