Do Spatial Characteristics Affect Housing Prices in Korea? : Evidence from Bayesian Spatial Models

This paper employs a Bayesian conditional autoregressive model to geographically analyze housing prices in Seoul, Korea from a demographic perspective. Spatial dependence patterns are detected between 424 administrative districts in Seoul, and the parameter estimation will be implemented via a Bayesian approach. We confirm that the proposed model with spatial heterogeneity presents superior performance than the other common spatial regression models. We also demonstrate that the proposed model offers the flexibility to resent various global spatial autocorrelation, and that the model adequately captures the model variablesʼ effect on housing prices.