Generalized Local Additive Spatial Smoothing (GLASS): A New Approach to Mitigating the Modifiable Areal Unit Problem in Local Regression Modeling

Addressing spatial non-stationarity and the Modifiable Areal Unit Problem (MAUP) in regression models is essential for accurate spatial analysis. This study introduces Generalized Local Additive Spatial Smoothing (GLASS), a technique that leverages spatial smoothing to adjust the spatial support of covariates to a specified target within local modeling. By doing so, GLASS aims to mitigate MAUP- related biases that arise from data aggregations that could affect the inference of local processes. Unlike traditional local regression approaches, which assume consistent spatial support across covariates, and often require prior aggregation of data before model calibration, GLASS endogenously aligns differing spatial supports during model calibration. Simulation results suggest that GLASS performs favorably compared to Geographically Weighted Regression (GWR), positioning it as a potential alternative for the spatial analysis of non-stationary relationships.