新疆各县市空间经济关联分析初步研究  

A preliminary study on Spatial Economic Association among Counties in Xinjiang

作　　者：陈斐[1] 高志刚[2] 苟中林 

机构地区：[1]南昌大学经济与管理学院,南昌330047 [2]新疆财经学院经济学系,乌鲁木齐830011 [3]乌鲁木齐市成人教育学院,乌鲁木齐830002

出　　处：《干旱区地理》2003年第3期274-280,共7页Arid Land Geography

基　　金：国家自然科学基金项目 ( 70 0 63 0 0 5 )资助

摘　　要：空间统计分析与GIS的有效集成 ,可以为确定、量化经济区域内的空间经济关联的性质和强度提供一个交互式的分析工具 ,结合区域分区 ,可以认识内在的局部空间经济关联模式及其动态变化 。Spatial autocorrelation means the self correlation or spatial dependence among observations of a geo referenced attribute. There are two different scales for spatial dependence: global indicators and local indicators. In this paper, the authors summarize a few spatial statistical analysis methods concerning about how to measure and identify spatial autocorrelation and spatial association firstly, then make a brief review about the integration of Spatial Statistical Analysis with GIS. Based on what has been done in this area, the authors point out that it is necessary and worthwhile to develop a user friendly statistical module combining spatial statistical analysis methods with GIS visual techniques in GIS directly, and provide an example to illustrate how this can be implemented in Arcview using Avenue. To construct spatial proximity weight matrix is the first step. A two dimensional matrix can be expressed as a one dimensional array by using the 'List' class. In this paper, we use a spatial proximity list table to represent spatially adjacent relations among different regional units. We take Xinjiang Uyger Autonomous Region as research area, and utilize mean Growth Rate of GDP(〔1978～1990, 1991～1999〕) in different counties, then calculate global MC and local MC based on those data, and illustrate the usefulness of that module in identifying the characteristic and significance of spatial association among observed locations over space. According to analytical results, there is a significant positive spatial autocorrelation between mean growth rates of GDP over 87 counties in Xinjiang, either in 1978～1990, or 1991～1999. We also investigate the spatial association between core counties and adjacent counties by computing the Local Moran and Geary Statistics at the county level. With the use of a conditional randomization or permutation approach, we can identify some different types of significant local spatial association based on the analysis of different counties. As a results, insight into the

关 键 词：新疆 空间经济关联 区域经济 GIS 空间统计分析 

分 类 号：F127[经济管理—世界经济]