Abstract:Summary Generalised linear mixed regression models are fundamental in statistics. Modelling random effects that are shared by individuals allows for correlation among those individuals. There are many methods and statistical packages available for analysing data using these models. Most require some form of numerical or analytic approximation because the likelihood function generally involves intractable integrals over the latents. The Bayesian approach avoids this issue by iteratively sampling the full conditional distributions for various blocks of parameters and latent random effects. Depending on the choice of the prior, some full conditionals are recognisable while others are not. In this paper we develop a novel normal approximation for the random effects full conditional, establish its asymptotic correctness and evaluate how well it performs. We make the case for hierarchical binomial and Poisson regression models with canonical link functions, for hierarchical gamma regression models with log link and for other cases. We also develop what we term a sufficient reduction (SR) approach to the Markov Chain Monte Carlo algorithm that allows for making inferences about all model parameters by replacing the full conditional for the latent variables with a considerably reduced dimensional function of the latents. We expect that this approximation could be quite useful in situations where there are a very large number of latent effects, which may be occurring in an increasingly 'Big Data' world. In the sequel, we compare our methods with INLA, which is a particularly popular method and which has been shown to be excellent in terms of speed and accuracy across a variety of settings. Our methods appear to be comparable to theirs in terms of accuracy, while INLA was faster, for the settings we considered. In addition, we note that our methods and those of others that involve Gibbs sampling trivially handle parameters that are functions of multiple parameters, while INLA approximations do not. Our primary illustration is for a three‐level hierarchical binomial regression model for data on health outcomes for patients who are clustered within physicians who are clustered within particular hospitals or hospital systems.

A Gentle Introduction to Bayesian Hierarchical Linear Regression Models

Approximate inferences for Bayesian hierarchical generalised linear regression models

Hierarchical Bayesian nonparametric models with applications

Bayesian Analysis for Social Science Research

Variational Bayesian hierarchical regression for data analysis

Hierarchical approaches for flexible and interpretable binary regression models

Bayesian estimation for longitudinal data in a joint model with HPCs

Non-Gaussian Normative Modelling With Hierarchical Bayesian Regression

Bayesian Nonparametric Modeling for Multivariate Ordinal Regression

Gibbs sampling for a Bayesian hierarchical general linear model

Learning Bayesian networks probabilities from longitudinal data

Fitting growth curve models in the Bayesian framework

Leveraging single-case results to Bayesian hierarchical modelling

A Bayesian approach to model individual differences and to partition individuals: case studies in growth and learning curves

Multinomial Models with Linear Inequality Constraints: Overview and Improvements of Computational Methods for Bayesian Inference

A multiscale Bayesian nonparametric framework for partial hierarchical clustering

Hierarchical Stochastic Model in Bayesian Inference: Theoretical Implications and Efficient Approximation

A Bayesian Model for Forecasting Hierarchically Structured Time Series

Bayesian analysis of linear regression models with autoregressive symmetrical errors and incomplete data

Learning Bayesian networks with heterogeneous agronomic data sets via mixed-effect models and hierarchical clustering

A Fully Nonparametric Modelling Approach to Binary Regression