Poisson M-Quantile geographically weighted regression on disease mapping

A new approach to ecological analysis on disease mapping is introduced: a semi-parametric approach based on M-quantile models. We define a Poisson M-Quantile spatially structured model. The proposed approach is easily made robust against outlying data values for covariates. Robust ecological disease mapping is desirable since covariates at area level usually present measure-type error. We consider a spatial structure in the model in order to extend the M-quantile approach to account for spatial correlation between areas using Geographically Weighted Regression (GWR). Differences between M-quantile and usual random effects models are discussed and the alternative approaches are compared using the Scottish Lip cancer example.