| The formation of urban housing price is the result of the combination of multiple factors.Quantitative analysis of the spatial distribution law of housing price and its relationship with influencing factors is helpful to deeply explore and grasp the internal law and formation mechanism of housing price,and to provide reference for government departments to make relevant decisions.Based on location theory and urban spatial structure theory,this paper combines the main methods of the existing research on the spatial differentiation of residential prices,using exploratory spatial data analysis method and spatial interpolation method to quantitatively analyze and visualize the spatial differentiation characteristics of residential prices in the main urban area of Xuzhou;and use OLS model and GWR model to quantify the role of the factors affecting the spatial differentiation of residential prices.The main contents and conclusions of the study are as follows:(1)With the support of GIS,use the exploratory spatial data analysis method to conduct normal distribution test,global trend analysis and spatial autocorrelation analysis on the residential price data in the study area.The study shows that the house price data in the main urban area of Xuzhou city follows a normal distribution after log transformation,and there is a U-shaped trend in both X and Y directions.The residential price in the study area shows a significant spatial positive correlation globally,and there are both high-value and low-value aggregation and a small number of heterogeneous points in the local area.The Kriging interpolation method was used to spatially interpolate the housing price data in the study area to generate a continuous housing price surface.The spatial differentiation model of the housing price in the study area was analyzed and summarized as a multi-core spatial structure.The price of housing decreases from the center to the outer layer,and its change is closely related to the location characteristics of the housing.(2)Process and quantify the collected residential price data,residential community attribute data,community samples and spatial data of influencing factors in the research area,and use GIS software to establish a residential sample data layer and attribute database in the research area,combined with the research area Planning plan and development status,select suitable influencing factors from the data of residential sample points to construct the OLS model and GWR model for studying residential prices and influencing factors in the area.Through comparative analysis and verification of the model results,it is found that the fit of the GWR model is better than that of the OLS model when studying data that exhibits spatial non-stationarity.Using the GIS software to draw the regression coefficient distribution map of the GWR model,the analysis shows that the housing price is affected by factors such as the age of the house,the greening rate,the bus stop and other factors.It shows a single positive or negative effect,and the effect is relatively stable in space.Factors such as large-scale shopping malls and park squares also have a single negative correlation with residential prices,but the size of their effect changes with the change of spatial location.The influence of factors such as urban centers,hospitals,primary and secondary schools on housing prices varies with location as a significant negative correlation in most areas and a positive effect in individual heterogeneous areas.The way in which subway stations affect residential prices varies greatly with location,manifested in the promotion effect in the southeast of the region and the suppression effect in the west.The influence of universities and scenic spots on the spatial differentiation of housing prices in regions has an abnormal effect that does not match prior knowledge.Therefore,it is determined that the overall performance is not significant,and the selection of specific items within such factors may be biased. |
PDF Full Text Request