Local Linear Estimation for Covariate‐Dependent Coefficients Model in Disease Mapping
Abstract
ABSTRACT Spatial regression effects may depend on covariates in disease‐mapping models. For example, the association between infectious disease incidence and risk factors may vary according to climatic factors, such as time, season, and temperature. In this study, we aim to develop a spatial model that accounts simultaneously for the spatial and varying effects of covariates believed to modulate the spatial association. We employ a local linear estimation method to estimate covariate‐dependent coefficients in a model for excess zero counts. The local linear estimator effectively smooths covariate‐dependent coefficient estimation. Comprehensive simulation studies were conducted to evaluate the performance of the local linear estimators, and reported dengue cases in villages in Kaohsiung City from January 2014 to December 2015 are used to inform our proposed method for practical use in real‐world applications.
// Source
Authors: Yexuan Jiang, Pei‐Sheng Lin, Jun Zhu, Feng‐Chang Lin
Institutions: Chicago Department of Public Health, University of North Carolina at Chapel Hill, University of Illinois Chicago, University of Wisconsin–Madison, National Health Research Institutes, National Chung Cheng University, University of Wisconsin System