Spatial Statistics
The first law of geography states that “everything is related to everything else, but near things are more related than distant things.” Spatial statistics is a form of statistical analysis in which the observations and/or spatial locations for which the observations are made are modeled as random variables. There are three primary subfields of spatial statistics which are distinguished by data type: (i) geostatistics, (ii) lattice (or areal) processes, and (iii) point processes. Geostatistics refers to point-referenced data where it is conceptually feasible to make observations at all possible sites within the continuous spatial domain. Lattice processes are counts or spatial averages of a quantity over a finite number of non-overlapping subregions of a larger spatial domain. Lastly, point processes are the arrangement of a countable number of spatial locations within a spatial domain. Spatial statistics aims to quantify, account for, and leverage the dependencies in each of these data types.