LAUSR

Abstract This article contributes to the recent literature in the spatial diffusion pattern of house prices by constructing spatial-temporal model to test the spatial relationships of 70 housing markets in China. Drawing on the spatial-temporal method proposed by Holly, a multiregional spatial-temporal model suitable for China is formulated which are applied to study the spatial-temporal diffusion effect of urban housing prices in China. Firstly, the diffusion mechanism of urban housing prices is analyzed from theoretical perspective, and four cities are assumed as dominant cities, namely, Beijing, Shanghai, Shenzhen and Sanya. Secondly, the spatial linkage of the housing prices of 70 large and medium-sized cities from 2007 to 2017 is examined by calculating the global Moran's I index. Thirdly, a spatial-temporal model is constructed to empirically analyze the diffusion effects of the housing prices of 70 sample cites. The study concludes that 1) significant global spatial autocorrelation is observed among the housing prices in the 70 cites; 2) empirical analysis shows that Beijing, Shanghai, Shenzhen and Sanya are the dominant cities of the 70 cities; and 3) the spatial effects of the four dominant cities on other cities differ in areas, degrees and duration.