Spatial Cluster Pattern and Influencing Factors of the Housing Market: An Empirical Study from the Chinese City of Shanghai

Infrastructure and amenities have an evident effect on differentiated urban structures and house prices. However, few studies have taken into account the spatial heterogeneity of large-scale urban areas. Regarding this issue, the present study proposes a novel spatial framework to quantify the impacts of built environment factors on the housing market. We aim to answer: how does a specific factor impact house prices across different spatially autocorrelated neighbourhood clusters? The city of Shanghai, the economic centre of China, is examined through the transaction data from the China Real-estate Information Center (CRIC) are analysed. Firstly, spatially autocorrelation clusters were explored to identify high/low housing prices in concentrated areas in Shanghai. Secondly, using the development-scale house prices as the dependent variable, we employed ordinary least squares (OLS) linear regression and geographically weighted regression (GWR) models to examine the impact of built environment facilities on the house prices across these spatial autocorrelation regions and Shanghai more generally. The results suggest the following: (1) There are significant spatially autocorrelated clusters across Shanghai, with high-value clusters concentrated in the city core and low value concentrated in the suburban fringes; (2) Across Shanghai and its spatially autocorrelated clusters, transportation accessibility and service amenities factors can affect house prices quite differently, especially when focusing on the city centre and the suburban areas. Our results highlight the importance of optimising the city’s polycentric structural framework to foster a more balanced regional development. Differentiated approaches to the distribution of public service facilities should be adopted to address the diverse needs of residents across various regions.