Geostatistical Air Pollution Indexes in Spatial Hedonic Models: The case of Madrid, Spain

Recent work has shown how spatial econometrics can be applied to a housingvalue hedonic equation that includes air pollution variables. In this paper we propose a Spatial Durbin Model that incorporates an Air Pollution Index, instead of one or two pollution variables, and considers spatial and non-spatial endogeneity jointly. We do, however, assume both possibilities for the index: exogenous and endogenous. Unlike most empirical studies in the literature, we have focused our empirical analysis on a European city, Madrid. In order to do so, we constructed a massive data base in 2009 that includes almost 12,000 dwellings.