Spatial Analysis Based On Second-Hand Housing Price

In recent years,Kunming has been expanding its urban area by virtue of its geographical advantages and becoming a third tier city in the world.Meanwhile,the real estate industry develops rapidly.The house price in Kunming has been in a state of continuous and stable growth,which has become a hot issue.There are many business districts in the main urban area of Kunming.With the opening of subway,the completion of large shopping plaza and the opening of branch schools in key schools,the old urban area in the first ring of Kunming no longer has absolute geographical advantages.The housing price in Kunming has changed both in overall level and spatial distribution pattern,which needs to be analyzed to provide necessary information support for the formulation of new urban planning.The research object in the paper is the second-hand housing price of Kunming.The software includes ArcGIS 10.5,SPSS 20.0 and GWR 4.0.Research methods include exploratory spatial data analysis and geostatistical analysis.Finally,Kriging interpolation image is created,and HPM model and half parameter GWR Model are established In the research:The average price of second-hand houses in the main urban area of Kunming is13836元/8)~2.The average price of community is mainly between 10000-17000元/8)~2.The housing price shows a strong positive correlation in the study area as a whole,and also shows an obvious clustering distribution in some areas.House prices in the East-West upward trend shows obvious change,South-North upward change is relatively slow.Kriging results show that the housing price presents an irregular ring structure in space.The two ring centers are located in Cuihu Lake and Dian Lake respectively.But the price in the northwest and southeast is relatively low.The significant variable include Age,Volume,Property,Carport,Line,Lake,CBD,Park,SC,PS and JS.The insignificant variable include Green,FT,Sub,Bus and Hosp.Finally,the semiparametric GWR Model of housing price is fitted for the first time.According to the comparison model fitting results,GWR Model is significantly better than HPM model.There is spatial nonstationary in housing price and its influencing factors.