研究了空间统计中最常用的局部空间自相关指标:Moran指数和G系数.基于模拟的空间区域,设计一些有代表性的空间聚集方案进行计算,比较两种指标的探测结果,得出局部G系数优于局部Moran指数的结论.
Local Moran index and Getis-Ord local G are the mostly common indices used for the analysis of lo- spatial autocorrelation. The two indices are compared based on simulated space by designing several spatial congregation schemes and their characters in finding the local spatial autocorrelation are analyzed. It is concluded that Getis-Ord local G is better than local Moran index.