Spatial Global Autocorrelation

This lecture introduces the concept of spatial autocorrelation, focusing on the determination of spatial weighting schemes and the interpretation of Moran's I as a regression coefficient. It covers the Geary contiguity ratio, second-order analysis of stationary point processes, and the significance of spatial autocorrelation through random permutations. The lecture also explores the visualization of spatial structures, the Moran scatterplot for local instability assessment, and the integration of exploratory spatial data analysis (ESDA) with geographic information systems (GIS). Practical aspects such as defining neighborhood criteria for point and polygon objects, creating spatial weighting schemes, and assessing spatial connectivity are discussed.