Extraction of Continuous and Discrete Spatial Heterogeneities: Fusion Model of Spatially Varying Coefficient Model and Sparse Modelling

Geospatial phenomena often have spatial heterogeneity, which is caused by differences in the data generation process from place to place. There are two types of heterogeneity: continuous and discrete, there has been much discussion about how analyze one type heterogeneity. Although geospatial can both heterogeneities, previous studies not sufficiently discussed consider these different heterogeneity simultaneously detect them separately, may lead biased estimates wrong interpretation phenomena. This study proposes a new approach for analysis with heterogeneities combining eigenvector filtering-based spatially varying coefficient (ESF-SVC) model, assumes generalized lasso (GL) estimation, discrete ESF-GL-SVC model. The performance was evaluated through experiments based on Monte Carlo simulation confirms that showed better estimating coefficients than models. application apartment rent outputs result smallest BIC value, estimated depict dataset. Reasonable were using ESF-GL-SVC, although some ESF-SVC not.