A Modified Gibbs Sampler on General State Spaces

TL;DR

This paper introduces a modified Gibbs sampler designed for general state spaces, demonstrating improved statistical efficiency without compromising convergence quality through theoretical analysis and practical examples.

Contribution

The paper proposes a novel modification to the Gibbs sampler that enhances efficiency in general state spaces, supported by theoretical and empirical validation.

Findings

Significant efficiency gains in sampling performance.

Maintains convergence quality comparable to standard Gibbs sampler.

Effective in models like Normal-Normal and Bayesian random effects.

Abstract

We present a modified Gibbs sampler for general state spaces. We establish that this modification can lead to substantial gains in statistical efficiency while maintaining the overall quality of convergence. We illustrate our results in two examples including a toy Normal-Normal model and a Bayesian version of the random effects model.