Abstract: Machine learning models are commonly used in applications with an objective of prediction. The complicated algorithms of many of these models, however, make them difficult to interpret. Methods have been proposed to provide insight into these "black-box" models, but there is little research that focuses on the situation when functional data are used as model inputs.
Abstract: Generalized linear mixed models (GLMMs) are often used to analyze non-Gaussian data arising from different studies. In Bayesian GLMMs, the commonly used improper priors may yield undesirable improper posterior distributions. Here we consider the popular improper uniform prior to the regression coefficients and several proper or improper priors including the widely used gamma and power priors on the variance components of the random effects. We derive necessary and sufficient conditions for posterior propriety for Bayesian binomial and Poisson GLMMs.
John L Eltinge, Assistant Director for Research and Methodology, U.S. Census Bureau
Title: Six Crises One Dozen Opportunities in Public-Stewardship Statistics
Lecturer with the Department of Economics, Mathematics & Statistics, Birkbeck, University of London
Title: Bayesian semiparametric modelling of covariance matrices for multivariate longitudinal data