Spatial Analysis Using Big Data

5.2. Spatial econometric models

5.2.1. Spatial lag model and spatial error model

In spatial econometrics, there are two representative models that are used to consider spatial autocorrelation (or dependence) among data—spatial lag model (SLM), which considers it as autocorrelation among observed variables, and spatial error model (SEM), which considers it as autocorrelation of error terms.

The SLM models the equilibrium outcome resulting from spatial and/or social interactions (Anselin, 2002; Brueckner, 2003). ² Although we cannot observe the actual process of such interactions from cross-sectional spatial data, it is possible to model the correlation structure in the equilibrium resulting from the interactions. Hence in contrast to geostatistical ...