Abstract
In this paper, we extend the Bayesian methodology introduced by Beamonte et al. (Stat Modelling 8:285–311, 2008) for the estimation and comparison of spatio-temporal autoregressive models (STAR) with neighbourhood effects, providing a more general treatment that uses larger and denser nets for the number of spatial and temporal influential neighbours and continuous distributions for their smoothing weights. This new treatment also reduces the computational time and the RAM necessities of the estimation algorithm in Beamonte et al. (Stat Modelling 8:285–311, 2008). The procedure is illustrated by an application to the Zaragoza (Spain) real estate market, improving the goodness of fit and the outsampling behaviour of the model thanks to a more flexible estimation of the neighbourhood parameters.
Similar content being viewed by others
Notes
The price distribution usually presents right skewness, so the dependent variable is transformed logarithmically in order to increase the degree of normality of the data.
References
Basu S, Thibodeau TG (1998) Analysis of spatial correlation in house prices. J Real Estate Finance Econ 17(1):61–85
Beamonte A, Gargallo P, Salvador M (2008) Bayesian inference in STAR models using neighborhood effects. Stat Modelling 8(3):285–311
Can A, Megbolugbe I (1997) Spatial dependence and house price index construction. J Real Estate Finance Econ 14(1):203–222
Dubin RA (1988) Estimation of regression coefficients in the presence of spatial autocorrelation error terms. Rev Econ Stat 70(3):466–474
Dubin RA (1992) Spatial autocorrelation and neighbourhood quality. Reg Sci Urban Econ 22(3):433–452
Dubin RA (1998) Predicting house prices using multiple listings data. J Real Estate Finance Econ 17(1):35–59
Gelfand AE, Ghosh SK, Knigth JR, Sirmans CF (1998) Spatio-temporal modeling of residential sales data. J Bus Econ Stat 16(3):312–321
Gelfand AE, Ecker MD, Knigth JR, Sirmans CF (2004) The dynamics of location in home price. J Real Estate Finance Econ 29(2):149–166
Pace RK, Gilley OW (1998) Generalising OLS and the grid estimator. Real Estate Econ 26(2):331–347
Pace RK, Barry R, Clapp JM, Rodriguez M (1998) Spatiotemporal autorregressive models of neighborhood effects. J Real Estate Finance Econ 17(1):14–33
Pace RK, Barry R, Gilley OW, Sirmans CF (2000) A method for spatial-temporal forecasting with an application to real estate prices. Int J Forecast 16(2):229–246
Spiegelhalter DJ, Best NG, Carlin BP, Van der Linde A (2002) Bayesian measures of model complexity and fit. J R Stat Soc B 64(4):583–639
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
In this Appendix, the posterior distribution of the parameters of model given in Eq. (1) and the algorithm used to calculate it are briefly described. From now on, we use the following notation. Given two random variables, X and Y, we will designate the marginal and conditioned densities of X and Y|X by [X] and [Y|X], respectively.
As in Pace et al. (1998, 2000) and Beamonte et al. (2008), we specify n 0 (max{m Tnet,t , m Snet,s } < n 0 < n), the number of observations to which we condition the likelihood function of model given in Eq. (1) in order to diminish the influence of the oldest transactions in the estimation of the parameters.
Let \( {\varvec{\uptheta}} = ({\varvec{\upalpha}}^{'} ,{\varvec{\upbeta}}^{'} ,{\varvec{\upphi}}^{'} ,\tau ,m_{T} ,\gamma ,m_{s} ,\lambda )^{'} \) be the model parameter vector and \( {\varvec{\upphi}} = (\phi_{T} ,\phi_{S} ,\phi_{ST} ,\phi_{TS} )^{'} \). Let \( {\mathbf{Y}}_{0} = (y_{1} , \ldots ,y_{{n_{0} }} )^{\prime } \)and \( {\mathbf{Y}}_{1} = (y_{{n_{0 + 1} }} , \ldots ,y_{n} )^{\prime } \)
Applying the Bayes Theorem, the posterior distribution of the parameters in Eq. (1) is given by:
where I A (x) = 1 if x ∈ A and 0 otherwise and ε 1(θ) is the (n−n 0)×1 vector that contains the last n−n 0 components of the errors vector ε(θ), where:
The algorithm used to draw a sample from Eq. (4) is based on Gibbs sampling and is similar to that described in Beamonte et al. (2008) with only two minor differences which affect the sampling of the neighbourhood parameters (m T , γ, m S , λ). To be precise,
-
(mT, mS)|α, β, ϕ, τ, γ, λ, Y, Z, X is a discrete distribution with support mTnet × mSnet and a probability function given by:
A Hasting-Metropolis step is used to draw a sample from this distribution, taking U D(m Tnet × m snet) as the transition density.
-
(γ, λ)|α, β, ϕ, τ, m T , m S , Y, Z, X is a continuous distribution with support (γmin,γmax)×(λmin,λmax) and a density function that is proportional to:
This distribution is not standard and we use a Hasting-Metropolis step to draw a sample from it, taking the uniform density in (γ min, γ max)×(λ min, λ max) as the transition density.
Rights and permissions
About this article
Cite this article
Beamonte, A., Gargallo, P. & Salvador, M. Analysis of housing price by means of STAR models with neighbourhood effects: a Bayesian approach. J Geogr Syst 12, 227–240 (2010). https://doi.org/10.1007/s10109-010-0115-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10109-010-0115-7