Official URL: http://www.geocomputation.org/2007/

In spite of literature on hedonic house price models is quite vast, much more efforts must be made to integrate and to model spatiotemporal data (see Cressie, 1993). Nevertheless, it is worth mentioning that excellent works have been done in recent years to deal with this phenomenon. For example, of paricular interest has been the research on panel data models (Holly et al., 2006) and spatiotemporal autoregressive (STAR) models (Pace et al., 1998). On the other hand, the development of geographically weighted regression (Brunsdon et al., 1998) has brought new insights into the understanding of spatial dynamics in econometric models. In this respect, we do believe that the incorporation of temporal data into the model and the subsequent development of a spatiotemporal version of GWR will naturally contribute to a better understanding of stochastic processes that vary and interact in space and time. This is presicely the aim of this study, that is, to develop a spatiotemporal version of the GWR technique, which would enable the forecasting of and eventually the interpolation of local parameters throughout time.