VALUING PRODUCT ATTRIBUTES USING SINGLE MARKET DATA - A COMPARISON OFHEDONIC AND DISCRETE CHOICE APPROACHES

This paper compares, via simulation, the performance of the multinomia l logit and hedonic models in estimating consumer preferences for prod uct attributes. We ascribe preferences over the attributes of houses t o a population of consumers, and, by having them bid for a set of hous es calculate equilibrium prices. The resulting data are used to estima te the two models. We find that the gradient of a linear Box-Cox hedon ic price function estimates marginal attribute bids at least as well a s a linear logit model, although the difference between the two is sma ll when some variables are not observed or are replaced by proxies. Th e logit model, however, outperforms the hedonic model in valuing non-m arginal attribute changes. This is true when the researcher knows the true form of consumers' utility functions and when the utility functio n must be approximated.