Two Approaches to Blending Spatial Weights and Temporal Weights in Calculating Spatiotemporal Autocorrelation

Abstract

Since Moran's seminal paper on an index to measure spatial autocorrelation among a set of geographic objects, spatial autocorrelation coefficients have been widely used in many research and application fields. In this paper, we provide detailed reasoning for indices for measuring spatiotemporal autocorrelation. We first briefly highlight the classic Moran's Index for measuring spatial autocorrelation. Next we introduce two methods that we used to blend spatial and temporal weights. Also the simulation experiment was adopted to evaluate the difference between the two methods.