arXiv subjects

Tavis Abrahamsen

Publications and source records attributed to Tavis Abrahamsen.

2 recordsLinked to original sources

arXiv2017

Fast Monte Carlo Markov chains for Bayesian shrinkage models with random effects

When performing Bayesian data analysis using a general linear mixed model, the resulting posterior density is almost always analytically intractable. However, if proper conditionally conjugate priors are used, there is a simple two-block Gibbs sampler that is geometrically ergodic in nearly all practical settings, including situations where $p > n$ (Abrahamsen and Hobert, 2017). Unfortunately, the (conditionally conjugate) multivariate normal prior on $β$ does not perform well in the high-dimensional setting where $p \gg n$. In this paper, we consider an alternative model in which the multivariate normal prior is replaced by the normal-gamma shrinkage prior developed by Griffin and Brown (2010). This change leads to a much more complex posterior density, and we develop a simple MCMC algorithm for exploring it. This algorithm, which has both deterministic and random scan components, is easier to analyze than the more obvious three-step Gibbs sampler. Indeed, we prove that the new algorithm is geometrically ergodic in most practical settings.

math.ST↗

arXiv2016

Convergence analysis of block Gibbs samplers for Bayesian linear mixed models with $p>N$

Exploration of the intractable posterior distributions associated with Bayesian versions of the general linear mixed model is often performed using Markov chain Monte Carlo. In particular, if a conditionally conjugate prior is used, then there is a simple two-block Gibbs sampler available. Román and Hobert [Linear Algebra Appl. 473 (2015) 54-77] showed that, when the priors are proper and the $X$ matrix has full column rank, the Markov chains underlying these Gibbs samplers are nearly always geometrically ergodic. In this paper, Román and Hobert's (2015) result is extended by allowing improper priors on the variance components, and, more importantly, by removing all assumptions on the $X$ matrix. So, not only is $X$ allowed to be (column) rank deficient, which provides additional flexibility in parameterizing the fixed effects, it is also allowed to have more columns than rows, which is necessary in the increasingly important situation where $p>N$. The full rank assumption on $X$ is at the heart of Román and Hobert's (2015) proof. Consequently, the extension to unrestricted $X$ requires a substantially different analysis.

math.ST↗