Revealing the influence of micro factors on urban land price and housing price is essential to improve the predicting precision of urban land price change and house price change and exercise effective macro-control over land market and real estate market.The aim of this study was to use the Geographically Weighted Regression model to quantitatively measure the impacts of local special natural features (lakes,rivers and mountains)on urban residential land and housing prices and their spatial differences,and to explain the cause from the aspect of impact factors.We found that the spatial distributions of Moran' s I,P value,R2,adjusted R2 and regression parameters of the testing models can characterize the spatially non-stationary relationship between natural features and residential land and housing prices accurately.When it comes to effect strengths,the average marginal residential land and housing price affected by lakes is 0.11 CNY/m2 and 0.52 CNY/m2,respectively,while that affected by lakes is 0.15 CNY/m2 on both residential land and housing prices.Lakes and rivers have more distinguished impacts on residential land and housing in areas with high prices according to the dynamic trends of their effects.Meanwhile,affected by the existence of surrounding commercial districts,anisotropic relationships between large lakes like East Lake and housing prices exist.Those mountains with a scattered distribution and low gradient,limited estate developments by the protected specific artificial amenities alike Yellow Crane Tower,show an insignificant influence.Rather than fitting with the spatial change trends of housing price,the spatial dynamics of residential land and housing prices are similar with that of regional trends and the change range is positively correlated with price.Rivers generate analogous impacts on both residential land and housing prices with similar marginal effects,but in two different spatial trends of gradients and spheres.
spatial non-stationarityresidential land pricehousing pricespecial natural featureslakerivermountainGWRWuhan
NETL
NSTL
万方数据
维普
