Research On The Spatial Variation And Impacting Factors Of Residential Land Price In Wuhan Based On GWR Model

Land price is a reflection of the economic role of land, affected by supply and demand on the land market, land prices are always in a state of flux. Judging from the level of space, land prices and their changes show with significant regional and geographical differences, that is the spatial differentiation of land prices. Researching on formation of spatial distribution of land prices, the factors and their regularity helps to give full play to the role of the price mechanism for regulating the land market, strengthen land price management and adjust the location of land supply and demand, so to implement the "smart planning" of land use as well as promoting the efficient allocation of urban land resources under the guidance of the city "smart growth" concept.Similar studies in the past mainly based on linear regression or hedonic price model, such models based on urban space homogeneous assumption which consider each factor’s contribution to residential land prices is equal in a different location inside the city. However, as the disparity distribution of natural and socio-economic resources inside the urban areas, factors affecting residential land prices about the marginal growth of residential land prices in different micro-locations should be variable. Geographically Weighted Regression (GWR) extends the traditional regression model framework. By assuming that the regression coefficient in linear regression model is a function of location of the observation point, the spatial characteristics of the data are included in the model, which creates the conditions for regression analysis of the relationship between spatial features, the model also optimizes the weights so can be used for partial estimates of space non-stability data. Thus, compared to conventional regression models, researching on the spatial distribution pattern of urban land prices with GWR model can be more in-depth understanding the spatial formation mechanism of urban land prices and their effect factors’mechanism.This article choose the urban residential land prices as study object and the study area is the main city of Wuhan, we analyze the spatial differentiation of residential land prices in Wuhan during2003-2012. We find that residential land prices in Wuhan still show a single central structure in general and show diminishing circle outward style in the core of the center; On this basis,we build GWR model based on the selected microscopic factors of residential land prices which include the site area, floor area ratio, universities, schools, hospitals, roads, subways, living goods market, water and parks to analyze the formation law of spatial distribution of residential land prices. The results show as follows:The main road volume rate, living goods market and educational resources have significantly geographical differences contribution to residential land prices. On the one hand, it shows urban residents’ preferences for ease of living, educational resources and traffic conditions when choose location of residence, on the other hand, it also indicate the unbalanced distribution of educational resources, life care market and traffic conditions in the study area.Compared with the traditional linear regression or hedonic price model, GWR model is more suitable for providing evidence from the micro perspective of location for residential space planning and land price update and also more in line with the concepts of urban planning in terms of "smart growth", it can provide theory and practice guidance to optimize the allocation of urban land resources.