A0333
Title: Scalable and robust regression models for continuous proportional data
Authors: Changwoo Lee - Duke University (United States) [presenting]
Benjamin Dahl - Duke University (United States)
Otso Ovaskainen - University of Helsinki (Finland)
David Dunson - Duke University (United States)
Abstract: Beta regression is used routinely for continuous proportional data, but it often encounters practical issues such as a lack of robustness of regression parameter estimates to misspecification of the beta distribution. An improved class of generalized linear models is developed, starting with the continuous binomial (cobin) distribution and further extending to dispersion mixtures of cobin distributions (micobin). The proposed cobin regression and micobin regression models have attractive robustness, computation, and flexibility properties. A key innovation is the Kolmogorov-Gamma data augmentation scheme, which facilitates Gibbs sampling for Bayesian computation, including in hierarchical cases involving nested, longitudinal, or spatial data. Robustness, ability to handle responses exactly at the boundary (0 or 1), and computational efficiency relative to beta regression are demonstrated in simulation experiments and through analysis of the benthic macroinvertebrate multimetric index of US lakes using lake watershed covariates.
