路网形态与住宅价格的多尺度空间关系研究——基于空间网络分析与多尺度地理加权回归模型

针对以往住宅价格影响因素分析中忽略城市路网形态参数的问题,该文基于空间网络分析,将不同搜索半径下道路接近度与穿行度纳入住宅价格特征变量,构建空间计量模型,探究成都市路网形态对住宅价格的驱动效应。结果表明:1)相比享乐回归模型和地理加权回归模型,多尺度地理加权回归模型揭示了不同变量的空间影响尺度,更适用于探究成都市路网形态对住宅价格的影响;2)路网的局部接近度、局部穿行度核心位于二环内,全局接近度核心主要分布在四环内,全局穿行度核心主要分布在交通干道上;3)不同搜索半径下对住宅价格具有显著驱动效应的接近度均为全局尺度变量,该效应在空间上相对平稳,而穿行度作为局部变量,其在空间上的影响尺度随搜索半径的变化而存在差异,具有一定的空间异质性,其中,路网接近度对住宅价格产生正向影响,其系数空间格局呈跨尺度相似特征,而穿行度产生负向影响,其系数空间格局呈跨尺度变异特征;4)各搜索半径下路网形态变量的差异可在一定程度上改变消费者对其他影响因素的偏好程度。

A Study on Multi-scale Spatial Relationship between Road Network Form and Housing Price Based on sDNA and MGWR

In view of the lack of taking the morphological parameters of urban road network into account in the influencing factors of housing price,based on the spatial design network analysis,the road closeness and betweenness under different search radii were included in the characteristic variables of housing price,and spatial econometric models was constructed to explore the driving effect of road network form on housing price in Chengdu.The results showed that compared with hedonic price model and geographically weighted regression model,multi-scale geographically weighted regression model can reveal the spatial impact scale of different variables,which is more suitable to explore the impact of road network form on housing price in Chengdu.The cores of local closeness,local betweenness and global closeness were mainly located in the one circle layer,while the core of global betweenness was mainly distributed on the main traffic roads.In particular,the closeness with significant effect on housing price under different search radii was a global scale variable,which was relatively stable in space,while the betweenness was a local variable,and its spatial influence scale varies with the change of search radius,which had a certain spatial heterogeneity.And the road network closeness had a positive impact on housing prices,and its coefficient spatial pattern showed the characteristics of cross-scale similarity,while the betweenness had a negative impact,and its coefficient spatial pattern showed the characteristics of cross-scale variation.The differences of road network morphological variables at various search radii could change consumers′preference for other influencing factors to a certain extent.