Geometric ergodicity of Gibbs samplers for linear latent models with GIG variance mixtures

TL;DR

This paper establishes the geometric ergodicity of Gibbs samplers for a broad class of linear latent non-Gaussian models with GIG variance mixtures, ensuring reliable convergence for statistical inference.

Contribution

It develops two complementary methods to prove geometric ergodicity across the entire GIG parameter space, including heavy-tail regimes and special cases like Student-t.

Findings

Markov operator is trace-class over large GIG parameter regions.

Geometric ergodicity is proven via drift and minorization in boundary regimes.

Numerical experiments demonstrate mixing efficiency and role of null-smallness constant.

Abstract

We study geometric ergodicity of the Gibbs sampler for linear latent non-Gaussian models (LLnGMs), a class of hierarchical models in which conditional Gaussian structure is preserved through generalized inverse Gaussian (GIG) variance-mixture augmentation. Two complementary routes to geometric ergodicity are developed for the marginal chain on the mixing variables. First, we show that the associated Markov operator is trace-class, and hence admits a spectral gap, over a large portion of the GIG parameter space. Second, for the remaining boundary and heavy-tail regimes, we establish geometric ergodicity via drift and minorization, subject to an explicit null-smallness condition that quantifies how the drift interacts with the null space of the observation operator. Together, these results cover the full GIG parameter space, including the normal-inverse Gaussian, generalized asymmetric Laplace, and Student-$t$ special cases. The geometric ergodicity of this chain underpins the consistency of Gibbs-based stochastic-gradient estimators for maximum likelihood estimation, and we provide conditions that make the required integrability checks transparent. Numerical experiments illustrate the theoretical findings, contrasting mixing efficiency across parameter regimes and probing the role of the null-smallness constant.