Non-parametric hedonic housing prices

The hedonic price function has long been a standard tool for modeling the price of complex commodities, such as housing. The theoretical basis of the model is sound and appealing, but applications often encounter difficulties. The results of hedonic models depend on inclusion of the right independent variables and the correct specification of the functional form. The functional form assumption is particularly difficult in the housing context because the hedonic price function summarizes not only consumer preferences and production technology, but also various quantities which are historically determined, difficult to measure, and not approachable by theory. In this paper, the functional form assumption is relaxed by estimating the hedonic price function as a General Additive Model (GAM). The GAM is considerably move general than conventional hedonic models and offers significant advantages in comprehensibility over other non-parametric procedures. The model is used to analyze the substantial decline in condominium house values in downtown Los Angeles during the 1980s-a period in which apartments lost about 40 per cent of their values.