Real estate market follows generally the evolution of the economic cycles. In Italy, it has been characterized by a period of expansion (2000–2006) and a period of regression (2007–2013), with a slow growth in the last years, which reflects essentially the European context. The empirical distribution of the number of normalized transactions (NNT), registered in the residential real estate market, provides a realistic picture of the Italian market and reveals several differences among the Northern, Central and Southern parts of the nation. Although time series analysis has been widely used to describe the temporal evolution of socio-economic variables, recently increasing attention has been also given to the study of the spatial distribution of these phenomena. The aim of the paper is to propose a geostatistical model for describing and predicting simultaneously the spatial and temporal evolution of the Italian NNT. After discussing the systematic variations of the NNT, the residuals at local level, which are characterized by a significant spatio-temporal correlation, are studied and modeled. The convenience of using a spatio-temporal model versus an only temporal model is analyzed.
File in questo prodotto:
Non ci sono file associati a questo prodotto.