A Study on Spatial Differentiation of Urban Land Prices Based on Geographically Weighted Regression Models

Abstract

Urban residential land price is related to the basic life of residents as well as the healthy and stable development of society. In this paper, the residential land price in QZ city was selected as the research object, and the residential land price data of 107 sample points in the built-up area were collected for the 10 years from 2013 to 2022, and the Exploratory Spatial Data Analysis (ESDA) method and Geostatistical Kriging Spatial Interpolation (GSI) method were combined to investigate the law of spatial heterogeneity of urban land price. Finally, general linear regression model and geographically weighted regression model are constructed to investigate the influencing factors of urban land price spatial differentiation by selecting suitable influencing factors. The sample data of residential land price in the study area basically conforms to normal distribution, and the land price shows a U-shape in the latitudinal direction, and in the horizontal direction, the land price shows a trend of decreasing from the city center to the periphery, and a “mono-core” spatial structure of decreasing from the center to the periphery. There are large regional differences in the contribution of plot ratio, distance from main roads, distance from hospitals and distance from schools to residential land price, among which the marginal effect of plot ratio on residential land price reaches 122.5385yuan/m², which reflects the importance of plot ratio to residential land price. The study provides a decision-making basis for government targeted land planning, urban infrastructure investment and real estate development.