Brian Neelon is an Associate Professor of Biostatistics in the Department of Public Health Sciences at the Medical University of South Carolina (MUSC) with a joint position at the Charleston VA Medical Center. He is also an Adjunct Associate Professor of Biostatistics at the University of North Carolina at Chapel Hill. His current work focuses on Bayesian methods for longitudinal, multivariate and spatiotemporal data. He has applied these methods to a wide range of clinical areas, including maternal and child health, cardiovascular disease, environmental heath, and health services utilization. Prior to his appointment at MUSC, he was an Assistant Professor of Biostatistics at Duke University, with joint positions at the Nicholas School of the Environment and the Center for Health Services Research at the Durham VA Medical Center. He currently serves as MPI on a VA Merit Award examining spatio-temporal trends in diabetes-related outcomes. Additionally, he is MPI for a grant from the National Library of Medicine to develop Bayesian statistical methods for shared medical decision making.
Talk: Bayesian Zero-Inflated Negative Binomial Regression Based on Pólya-Gamma Mixtures
This talk will be held via zoom. A link and password were sent to the SDS community via email on Wednesday morning.
Abstract: Motivated by a study examining spatiotemporal patterns in inpatient hospitalizations, we propose an efficient Bayesian approach for fitting zero-inflated negative binomial models. To facilitate posterior sampling, we introduce a set of latent variables that are represented as scale mixtures of normals, where the precision terms follow independent Pólya-Gamma distributions. Conditional on the latent variables, inference proceeds via straightforward Gibbs sampling. For fixed-effects models, our approach is comparable to existing methods. However, our model can accommodate more complex data structures, including multivariate and spatiotemporal data, settings in which current approaches often fail due to computational challenges. Using simulation studies, we highlight key features of the method and compare its performance to other estimation procedures. We apply the approach to a spatiotemporal analysis examining the number of annual inpatient admissions among United States veterans with type 2 diabetes. We also discuss recent applications to COVID-19 and social vulnerability, using Alabama and Arizona as case studies.