Content area
Full Text
The Journal of Real Estate Finance and Economics, 31:2, 155187, 2005# 2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands.A Spatio-Temporal Autoregressive Model
for Multi-Unit Residential Market Analysis*HUA SUNDepartment of Real Estate, School of Design and Environment, National University of Singapore, Singapore andCentre for Urban Economics and Real Estate, Sauder School of Business, University of British Columbia, CanadaE-mail: [email protected] TUDepartment of Real Estate, School of Design and Environment, National University of Singapore, SingaporeE-mail: [email protected] YUDepartment of Real Estate, School of Design and Environment, National University of Singapore, SingaporeE-mail: [email protected] splitting the spatial effects into building and neighborhood effects, this paper develops a two order spatiotemporal autoregressive model to deal with both the spatio-temporal autocorrelations and the heteroscedasticityproblem arising from the nature of multi-unit residential real estate data. The empirical results based on 54,282condominium transactions in Singapore between 1990 and 1999 show that in the multi-unit residential market, atwo order spatio-temporal autoregressive model incorporates more spatial information into the model, thusoutperforming the models originally developed in the market for single-family homes. This implies that thespecification of a spatio-temporal model should consider the physical market structure as it affects the spatialprocess. It is found that the Bayesian estimation method can produce more robust coefficients by efficientlydetecting and correcting heteroscedasticity, indicating that the Bayesian estimation method is more suitable forestimating a real estate hedonic model than the conventional OLS estimation. It is also found that there is atrade off between the heteroscedastic robustness and the incorporation of spatial information into the modelestimation. The model is then used to construct building-specific price indices. The results show that the priceindices for different condominiums and the buildings within a condominium do behave differently, especiallywhen compared with the aggregate market indices.Key Words: spatio-temporal autocorrelation, spatio-temporal model, heteroscedasticity, Gibbs Sampling,Bayesian, Singapore condominium market1. IntroductionLocation and time play an important role in explaining real estate prices. Locationmatters because of the spatial dependence between real estate prices and the spatialheterogeneity across prices.*This paper was presented at the SingaporeYHong Kong International Real Estate Research Symposium,
organized by the Department of Real Estate, National University of Singapore, from 18 to 19 July, 2003.156 SUN ET AL.Spatial dependence or spatial autocorrelation refers to the possible occurrence of theinterdependence among the observations viewed in a geographical space,...