Coarse-to-fine spatial GLMM for scalable prediction and multiscale analysis

TL;DR

This paper extends coarse-to-fine spatial modeling to generalized linear mixed models for count data, improving scalability and addressing degeneracy, with applications to COVID-19 analysis.

Contribution

It introduces CF-GLMM, a scalable coarse-to-fine spatial model for non-Gaussian data, expanding applicability beyond Gaussian responses.

Findings

CF-GLMM effectively predicts spatial count data.

The method extracts multiscale spatial features.

Application to COVID-19 demonstrates practical utility.

Abstract

Although a recent study suggested that coarse-to-fine learning provides a fast and flexible framework for large-scale spatial process modeling, the method was originally developed for Gaussian responses, limiting its applicability. To address this limitation, we extended the coarse-to-fine spatial modeling (CFSM) framework to accommodate spatial generalized linear mixed models (GLMMs), with a particular focus on count data. The resulting model, referred to as CF-GLMM efficiently addresses the degeneracy problem often encountered in conventional spatial GLMMs. The performance of the proposed CF-GLMMs was evaluated in terms of spatial prediction and multiscale feature extraction via Monte Carlo experiments. Finally, we applied the proposed method to the analysis of coronavirus disease 2019 (COVID-19). The proposed method is implemented in an R package spCF (https://cran.r-project.org/web/packages/spCF/).