空间自相关的可塑性面积单元问题效应

可塑性面积单元问题(modifiable areal unit problem,MAUP) 效应是对空间数据分析结果产生不确定性影响的主要原因之一,在空间自相关分析中也不例外.本文分别利用网格模拟数据和中国人均GDP实例数据为数据源,以全局Moran's I 系数来探究空间自相关统计中的MAUP效应,分析结果表明,变量的空间自相关程度依赖于空间的粒度大小与单元的划分方法,但空间单元的变化与自相关性并不存在某种函数关系.因此,在进行空间自相关研究时必须选择合适的地理单元的粒度大小和分区.最后本文给出一种基于地统计内插方法来降低MAUP对空间自相关分析影响.

Effect of MAUP in Spatial Autocorrelation

This research investigated the role of modifiable area unit problem (MAUP) in the spatial autocorrelation on data of per capita GDP of China and grid simulated data. The global Moran's I coefficient was used to explore the MAUP effect of spatial autocorrelation statistics. The results showed that the degree of spatial autocorrelation of variables depends on the size of spatial particles and zone designing methods. However, there is no determined function relationship between the change of spatial units and the autocorrelation. So, how to choose the appropriate size of geographical unit particles and zone designing are very important in the research of spatial autocorrelation. Finally, the study presented an approach based on geostatistical interpolation to decrease MAUP effect in spatial autocorrelation analysis.