Scalable Bayesian computation for crossed and nested hierarchical models

TL;DR

This paper introduces scalable Bayesian sampling algorithms for crossed and nested hierarchical models, significantly improving computational efficiency while maintaining accuracy, applicable to large datasets in applied sciences.

Contribution

It develops novel scalable algorithms for Bayesian hierarchical models, leveraging sparse structures and local centering, with theoretical analysis and practical validation.

Findings

Algorithms scale linearly with data size

Significant performance improvements over Hamiltonian Monte Carlo

Effective in real-world applications like electoral and real estate data

Abstract

We develop sampling algorithms to fit Bayesian hierarchical models, the computational complexity of which scales linearly with the number of observations and the number of parameters in the model. We focus on crossed random effect and nested multilevel models, which are used ubiquitously in applied sciences. The posterior dependence in both classes is sparse: in crossed random effects models it resembles a random graph, whereas in nested multilevel models it is tree-structured. For each class we identify a framework for scalable computation, building on previous work. Methods for crossed models are based on extensions of appropriately designed collapsed Gibbs samplers, where we introduce the idea of local centering; while methods for nested models are based on sparse linear algebra and data augmentation. We provide a theoretical analysis of the proposed algorithms in some simplified settings, including a comparison with previously proposed methodologies and an average-case analysis based on random graph theory. Numerical experiments, including two challenging real data analyses on predicting electoral results and real estate prices, compare with off-the-shelf Hamiltonian Monte Carlo, displaying drastic improvement in performance.