关键词: GIS, GWR模型, 住宅地价, 时空演变, 江苏省

Abstract:

Based on the analysis of general variation characteristics of land price in Jiangsu Province, this article attempts to build a GWR model for residential land price in urban areas designed to study the factors responsible and the spatial variation characteristics in 1997, 2005 and 2008. The study reveals the degrees of the influential factors and regional differences on land prices, and explores the theoretical research of spatio-temporal variation laws of the factors that contribute to urban land price. The results show: 1) the distance from a central city to residential lands produces negative impacts on the land price, and the average of the absolute value of the regression coefficient increases gradually over the three years with a gradual expansion of high value areas; 2) GDP is positively correlated to residential land price, and the regions where GDP has the most powerful influences on residential land price shift gradually to the northern part of Jiangsu Province; and the number of areas with high value reduces gradually; 3) urbanization rate shows a positive correlation with residential land price, in the sense that from spatial perspective. In the areas along Changjiang River which link the central and southern parts of Jiangsu Province, urbanization rate exerts increasing impacts on residential land price, whereas, from temporal perspective, urbanization rate produces decreasing effects on residential land price; 4) outbound traffic condition shows a positive impact on residential land price, and the regression coefficient in the southern part is fairly high but gradually decreases from the south to the north; from temporal perspective, its influence shows an upward trend; 5) the number of employees is positively correlated to residential land price; the regression coefficient in the southern part of Jiangsu Province is fairly high and the absolute value of regression coefficient gradually increases from the south to the north; besides, the number of employees produces more effects on the southern part than on the northern part of Jiangsu Province; 6) the regression coefficient of the total real estate investment in the northern part is fairly high, and it gradually declines from the northern part to the southern part; the total real estate investment is more influential to the residential land price of the northern part than to that of the southern part of Jiangsu Province.

Key words: GIS, GWR Model, residential land price, spatio-temporal variation, Jiangsu Province