New Approaches to Model Simulated Spatio-Temporal Moran\u27s Index

The Moran\u27s index is a statistic that measures spatial autocorrelation; it quantifies the degree of dispersion (or clustering) of objects in space. However, when investigating data over a general area, a single global Moran statistic may not give a sufficient summary of the spread, behavior, features or latent surfaces shared by neighboring areas; rather, by partitioning the area and taking the Moran statistic of each divided subareas, we can discover patterns of the local neighbors not otherwise apparent. In this paper, we present a simulation experiment where the local Moran values are computed and a time variable is added to a spatial Poisson point process. Changes in the Moran statistics over the neighboring areas are investigated and ideas on how to perform the analysis are proposed