Shen, Y; Karimi, K; (2015) Understanding the roles of urban configuration on spatial heterogeneity and submarket regionalisation of house price pattern in a mix-scale hedonic model: The case of Shanghai, China. In: Karimi, K and Vaughan, L and Sailer, K and Palaiologou, G and Bolton, T, (eds.) Proceedings of the 10th International Space Syntax Symposium (SSS10). Space Syntax Laboratory,The Bartlett School of Architecture, UCL (University College London): London, UK.

Preview

Text SSS10_Proceedings_098.pdf - Published Version Download (2MB) | Preview

Abstract

Hedonic model is a powerful tool to investigate the important factors featuring cities' house markets globally; but the development of a local regression approach named as 'Mixed Geographically Weighted Regression' (MGWR) has brought a new insight into urban studies in the field of house price modelling with a proper consideration of spatial heterogeneity, which enables urban planners to know valuable local and global information about elements that factor the property value. In this paper, we proposed a two-step framework to understand the spatial heterogeneity and submarket regionalisation via MGWR in a case study of Shanghai. In the first step, a mixed GWR hedonic model is adopted with the incorporation of globally fixed effect and local factors. In this manner, the influence surface of spatial configuration and land use settings on house price patterns are measured by controlling other elements. Compared with the outputs of ordinary least squares (OLS) analysis and purely local model, the results of specified model in this study are more promising, indicating that the spatial heterogeneity of house price is a complex urban system. In the second stage, a data-driven approach is adopted to detect the spatially homogeneous submarkets based on the coefficients of accessibility indices obtained from MGWR instead of the raw data. Comparison of a regular cluster analysis using housing structural information and the prosed approach is conducted. The results suggest that spatial accessibility and function accessibility are key factors in shaping spatially continuous housing sub-markets on multiple scales with other elements. The findings finally reveal that multi-level modelling procedures provide additional and useful insights into the varying relationships between spatial layouts and the housing price distribution and that urban configuration not only affects house price pattern but also influence shifting housing submarkets.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item