Coro Chasco-Yrigoyen, Department of Applied Economics
Universidad Autónoma de Madrid, Madrid, Spain, Fernando López-Hernández, Department of Quantitative and Computering Methods
Universidad Politécnica de Cartagena, Cartagena, Spain
SPACE-TIME REGRESSION MODELS APPLIED TO THE ESTIMATION OF HOUSING PRICES OF THE SPANISH PROVINCES (assigned to theme
Nowadays, there is an increasing interest in panel data models. It has also been extended to the field of spatial econometrics by means of the introduction of time dimension in purely spatial models. The spatial SUR models are an example of this kind of space-time specifications, in which neighborhood is considered as a contemporaneous phenomenon between groups of spatial units in a period of time. That is to say, they do not consider non-contemporaneous or lagged spatial interactions. The aim of this paper is to present a generalization of the spatial SUR models using a derivation of the diagonal spatial dependence coefficients matrix: a space-time dependence lower-diagonal coefficients matrix. Consequently, this space-time regression model includes both contemporaneous (instant) and non-contemporaneous spatial dependence terms. In this paper, we demonstrate the superiority of these specifications over the classic spatial SUR models to explain housing price of Spanish provinces.
 | Conference organized through conf-vienna
(copyright Gunther Maier) |