Research On The Spatial Variation And Effect Factors Of The Urban Residential Land Prices Based On Geographically Weighted Regression Model

Urban residential land is related to the healthy and stable development of basic living and social residents. By Researching on the Spatial Variation and Effect Factors of the Urban Residential Land Prices in small and medium-sized cities, which can be provide decision-making basis for the government’s land planning, urban infrastructure investment and real estate development.This paper analyzed space variation pattern rule, analyzed the different rules of residential land in the Qinzhou District of Tianshui City by combining GIS spatial analysis method, finally studied Tianshui City residential land price influence factors by establishing general linear regression model and the GWR Model. The main achievements of this paper are:(1) Residential land data accord with normal distribution, the residential land price in the latitudinal direction presented an inverted U-shaped, and longitude direction curve is a gradually decreasing from north to south of the parabola, land price curve values presented in the east-west direction low high low state of distribution, in the north-south direction is north to south. Residential land price spatial variation pattern is single kernel space structure by Ordinary Kriging interpolation, which is from the center outward circumference decreased: residential land price has the slower pace of decline in east-west direction, but the faster speed of the north and south.(2)This paper selects the main roads, schools, public transportation, hospital, service facilities, urban green space and volume rate as the influence factor of the Qinzhou District of Tianshui City residential land, through the establishment of the general linear regression model and the GWR model, analyzed the association between land price and its influencing factors by quantitative analysis. The order of the average marginal contribution on the residential land price from high to low is volume rate, the distance from bus stations, the distance from schools, the distance from living facilities, the distance from urban green space, the distance from hospitals, the distance from the main road. Volume rate, the distance from the main road, the distance from hospitals, the distance from schools have significantly geographical differences contribution to residential land prices.(3)The paper is concluded with that GIS tools make GWR regression results visualizing, clearly show the residential land price factors spatial differences, we can understand area planning efforts to control, traffic conditions, medical conditions, environmental conditions, etc. which can be provide decision-making basis for the government’s land planning, urban infrastructure investment and real estate development.