Spatial Econometric Study On The Price Of Second-hand Housing And Its Influencing Factors In Beijing City

Since the influence factors of urban housing price are complex and rather diverse,the spatial heterogeneity of housing price not only affect the housing prices in different cities,but even in different business districts within a city.In order to reflect the marginal influence differences of locational factor on the housing prices in different area,Geographically Weighted Regression model is used in this thesis with the housing price spatial data of sample points in Beijing to study on the influence factors of urban housing price and its spatial heterogeneity.This thesis selected a sample of second-hand houses within the sixth ring road of Beijing city to be the research target.Exploring spatial data analysis such as normal distribution test,global trend analysis,spatial autocorrelation analysis and direction variation analysis were conducted to explore the spatial regulation of housing prices on second hand houses.This thesis used the all-subsets regression method in order to filter 7 core influence factors needed to build the GWR model.Thus,the spatial interactions of housing prices of these sample points are put into consideration.Based on the regression result,this thesis explores the influence factors of the urban housing price and regional difference characteristics,the reason for spatial differentiation of housing price can also be explained.This thesis shows the following conclusions: 1)The housing price of second-hand houses in Beijing city is highly influenced by the ring of road in which the sample points are located.If the Forbidden City is regarded as the center of Beijing city,with the increase of the number of road loops from the 2nd ring road to the 6th ring road,the unit price of the house will show a radical downward trend.If the region within 6th ring road in the Beijing city is horizontally divided into the northern half part and the southern half part by the center of the city,it is found that among the sample points with the same distance to the Forbidden City,the housing price of the sample points in the northern half part is generally higher than the housing price of the sample points in the southern half part.2)Spatial autocorrelation and cluster effect exist in the unit price of the second-hand house in Beijing city.The sample points with high housing price cluster with other sample points with high housing price,and the sample points with low housing price cluster with other sample points with low housing price,which proves that the nearby sample points in two-dimensional space will affect each other's housing price level and have similar house unit price.3)Among the seven factors influencing the house pricing,accessibility to commercial service center is the key factor.The spatial differentiation of how this factor influences the unit price of the second-hand house in different areas within Beijing city is quite obvious.4)The Geographically Weighted Regression model better fits the influence factors of unit price of the second-hand house rather than the traditional linear regression model.Based on the conclusions above,this thesis suggests to the relevant governmental department as the followings: 1)Urban planning should be formulated according to the local conditions of the region to optimize the urban spatial structure.2)Targeted measures should be taken to control the housing price so as to improve the effectiveness of policies.