Research Of Spatial Distribution Characteristics And Influence Factors Of Housing Prices In Xi'an City Based On ArcGIS

Housing prices are closely related to the basic life of residents and the healthy and stable development of society.In view of the less research on the spatial distribution of housing price in the second line city of the west,the spatial distribution characteristics and the influencing factors of the housing price are studied,and the corresponding decision support is provided for the land planning,the urban infrastructure construction and the real estate development.From the perspective of geostatistics,taking the housing price data of 451 main cities in Xi'an as an example,the spatial autocorrelation,variation function and Kriging spatial interpolation are used in the empirical analysis to summarize the spatial structure characteristics of the urban housing price,and prove the spatial non-stability of the housing price.Then select the appropriate influencing factors to construct the general Hedonic characteristic price model and the GWR geographical weighted regression model to study the influence factors of the housing price in Xi'an.The main research results of this paper are as follows:(1)By means of geostatistics,it is found that the overall residential price of Xi'an has a positive spatial correlation and its overall Moran's I index is 0.333.In the isotropy,the optimal fitting model of housing price is the index model,and the model determination coefficient(R~2)is 0.974,but the variation degree in different directions is not the same.At present,the spatial distribution of housing prices in the main urban area of Xi'an is characterized by the low south high north,the East High West low,the existence of significant spatial variability,and there is an extreme center,two sub core groups,and two new groups.(2)The influence factors of residential prices in the main city of Xi'an are selected as commercial center,bus station,subway,road network density,hospital,school,park and water area.By establishing Hedonic characteristic price model and GWR geographic weighted regression model,the relationship between housing price and its influencing factors is quantitatively analyzed.The model results show that the fitting degree(R~2)of the GWR model is better than that of the Hedonic model.From the regression results of the GWR model,there is a great difference in the spatial influence of the various factors on the housing price.The influence of the business service center,the subway,the hospital,the school,the bus station and the water area on the housing price is more significant,and it has a strong effect on the housing price.Correlation,road network density has little effect on housing price,not the main influencing factor of housing price.(3)Using ArcGIS to establish the spatial database of housing price in Xi'an City,through the spatial analysis method,the spatial distribution characteristics of housing price and the regression results of GWR model are visualized and analyzed.The spatial structure of housing price in Xi'an and the spatial difference of influence factors are displayed directly,and the housing pattern of Xi'an city is also analyzed.Suggestions are put forward in the optimization.