基于多尺度地理加权回归模型的中国城市网络外部性的空间异质性比较研究  

An comparative study on spatial heterogeneity of China's urban network externality based on Multiscale Geographically Weighted Regression Model

作　　者：姚雯雯 赵新正[1,2,3,4] 马海涛 李培庆[1] YAO Wenwen;ZHAO Xinzheng;MA Haitao;LI Peiqing(College of Urban and Environmental Science,Northwest University,Xi'an 710127,China;Yellow River Institute of Shaanxi Province,Northwest University,Xi'an 710127,China;Shaanxi Institute of Provincial Resource,Environmental and Development,Xi'an 710127,China;Shaanxi Key Laboratory of Earth Surface System and Environmental Carrying Capacity,Xi'an 710127,China;Key Laboratory of Regional Sustainable Development Modeling,Institute of Geographic Sciences and Natural Resources Research,CAS,Beijing 100101,China)

机构地区：[1]西北大学城市与环境学院,西安710127 [2]西北大学陕西省黄河研究院,西安710127 [3]陕西省情研究院,西安710127 [4]陕西省地表系统与环境承载力重点实验室,西安710127 [5]中国科学院地理科学与资源研究所,区域可持续发展分析与模拟重点实验室,北京100101

出　　处：《地理研究》2024年第5期1073-1087,共15页Geographical Research

基　　金：国家自然科学基金项目(42371226)。

摘　　要：本文以全国A股上市公司及其分支机构数据构建的城市经济网络和以合著论文数据构建的城市知识网络为基础,借助多尺度地理加权模型与社会网络分析等方法,对全国240个地级及以上城市嵌入上述两类网络后在经济增长方面所受影响的空间异质性进行实证研究和比较。结果表明:(1)嵌入不同类型城市网络后对城市经济增长存在不同的促进作用和空间溢出差异。相比于嵌入城市经济网络,嵌入城市知识网络对于城市经济增长的促进作用更强;嵌入城市经济网络对于城市经济增长具有正向促进或负向抑制两种差异化影响且呈现出小尺度溢出特征,嵌入城市知识网络对于城市经济增长具有正向促进作用且呈现大尺度溢出特征。(2)不同规模城市嵌入两类城市网络后在经济增长方面所获得的收益存在较大差异。特大超大城市嵌入城市经济网络后有较大概率获得较高的经济收益,大城市和小城市分别在嵌入城市知识网络和城市经济网络后有较大概率获得高经济收益,中等城市嵌入两类城市网络后在获得高经济收益方面均无明显优势。(3)不同区域城市在嵌入两类城市网络后获得的经济收益呈现出总体上的区域性差异特征和局部地区的地理临近相似特征。东部沿海和西北地区城市从城市经济网络中获得的经济收益最高,与这些城市相临近的中西部地区城市获得的经济收益次之,内陆其他中西部地区城市获得的经济收益最低。城市嵌入城市知识网络后获得的经济收益呈现明显的由北向南梯度递减特征。Based on the urban economic network constructed from the data of China's A-share listed companies and their branches and the urban knowledge network constructed from the data of co-authored papers in China, this paper uses Multiscale Geographical Weighted Regression model (MGWR) and social network analysis methods to conduct an empirical study and comparison on the 240 cities at prefecture level and above, and to examine their spatial heterogeneity of economic growth in the above two types of networks. The results show that: (1) Embedding different types of urban networks has different effects on urban economic growth and spatial spillover. Compared to embedding urban enterprise networks, embedding inter-city knowledge network has a stronger promoting effect on urban economic growth. Embedding inter-city enterprise network has a dual effect of promoting and inhibiting urban economic growth, and exhibits significant small-scale spillover characteristics, while embedding urban knowledge networks universally promotes urban economic growth and exhibits large-scale spillover characteristics. (2) Cities of different scales obtain significantly different benefits in terms of economic growth after embedding in two types of urban networks. Mega cities with a population of less than 10 million are more likely to achieve high economic benefits after embedding in urban economic network. Large and small cities have a high probability of obtaining high economic benefits after embedding into urban knowledge networks and urban economic networks, respectively. Medium-sized cities have no significant advantage in obtaining high economic benefits after embedding into both types of urban networks. (3) Cities in China show regional differences in economic benefits obtained after embedding in both types of networks, as well as local geographic similarity in certain areas. Cities in the eastern coastal and northwestern regions receive the highest economic benefits from urban economic networks, followed by cities in the central and west

关 键 词：城市网络外部性 空间异质性 多尺度地理加权回归模型 城市经济网络 城市知识网络 

分 类 号：TU984[建筑科学—城市规划与设计]