Variational Inference for Generalized Linear Mixed Models Using Partially Noncentered Parametrizations
Authors:
- Linda S. L. TanLTLinda S. L. Tan
- David J. NottDNDavid J. Nott
Open Access
- 1 May 2013
- journal article
- Published by Institute of Mathematical Statistics in Statistical Science
- Vol. 28 (2) , 168-188
Abstract
The effects of different parametrizations on the convergence of Bayesian computational algorithms for hierarchical models are well explored. Techniques such as centering, noncentering and partial noncentering can be used to accelerate convergence in MCMC and EM algorithms but are still not well studied for variational Bayes (VB) methods. As a fast deterministic approach to posterior approximation, VB is attracting increasing interest due to its suitability for large high-dimensional data. Use of different parametrizations for VB has not only computational but also statistical implications, as different parametrizations are associated with different factorized posterior approximations. We examine the use of partially noncentered parametrizations in VB for generalized linear mixed models (GLMMs). Our paper makes four contributions. First, we show how to implement an algorithm called nonconjugate variational message passing for GLMMs. Second, we show that the partially noncentered parametrization can adapt to the quantity of information in the data and determine a parametrization close to optimal. Third, we show that partial noncentering can accelerate convergence and produce more accurate posterior approximations than centering or noncentering. Finally, we demonstrate how the variational lower bound, produced as part of the computation, can be useful for model selection.Keywords
All Related Versions
This publication has 41 references indexed in Scilit:
- Variational Approximation for Mixtures of Linear Mixed ModelsJournal of Computational and Graphical Statistics, 2014
- The Discrete Infinite Logistic Normal DistributionBayesian Analysis, 2012
- Gaussian Variational Approximate Inference for Generalized Linear Mixed ModelsJournal of Computational and Graphical Statistics, 2012
- To Center or Not to Center: That Is Not the Question—An Ancillarity–Sufficiency Interweaving Strategy (ASIS) for Boosting MCMC EfficiencyJournal of Computational and Graphical Statistics, 2011
- Explaining Variational ApproximationsThe American Statistician, 2010
- Bayesian inference for generalized linear mixed modelsBiostatistics, 2009
- A comparison of Bayesian and likelihood-based methods for fitting multilevel modelsBayesian Analysis, 2006
- Maximum Likelihood for Generalized Linear Models with Nested Random Effects via High-Order, Multivariate Laplace ApproximationJournal of Computational and Graphical Statistics, 2000
- Parameter Expansion for Data AugmentationJournal of the American Statistical Association, 1999
- A likelihood-based method for analysing longitudinal binary responsesBiometrika, 1993