Uniform ergodicity of the Particle Gibbs sampler

Abstract

The particle Gibbs (PG) sampler is a systematic way of using a particle filter within Markov chain Monte Carlo (MCMC). This results in an off-the-shelf Markov kernel on the space of state trajectories, which can be used to simulate from the full joint smoothing distribution for a state space model in an MCMC scheme. We show that the PG Markov kernel is uniformly ergodic under rather general assumptions, that we will carefully review and discuss. In particular, we provide an explicit rate of convergence which reveals that: (i) for fixed number of data points, the convergence rate can be made arbitrarily good by increasing the number of particles, and (ii) under general mixing assumptions, the convergence rate can be kept constant by increasing the number of particles superlinearly with the number of observations. We illustrate the applicability of our result by studying in detail two common state space models with non-compact state spaces.

References in corpus (3)

• The pseudo-marginal approach for efficient Monte Carlo computations
• Particle Gibbs with Ancestor Sampling
• Efficient Bayesian Inference for Switching State-Space Models using Discrete Particle Markov Chain Monte Carlo Methods

Cited by in corpus (16)

• Particle Filters and Data Assimilation
• Nonlinear System Identification: Learning while respecting physical models using a sequential Monte Carlo method
• Importance sampling type estimators based on approximate marginal MCMC
• Sequential Monte Carlo as Approximate Sampling: bounds, adaptive resampling via $\infty$-ESS, and an application to Particle Gibbs
• Particle ancestor sampling for near-degenerate or intractable state transition models
• Interacting Particle Markov Chain Monte Carlo
• On particle Gibbs Markov chain Monte Carlo models
• Perfect simulation using atomic regeneration with application to Sequential Monte Carlo
• Scalable Monte Carlo inference for state-space models
• A Sharp First Order Analysis of Feynman-Kac Particle Models
• Particle Gibbs algorithms for Markov jump processes
• Learning dynamical systems with particle stochastic approximation EM
• On Unbiased Score Estimation for Partially Observed Diffusions
• Markov Interacting Importance Samplers
• NEO: Non Equilibrium Sampling on the Orbit of a Deterministic Transform
• Stability of Conditional Sequential Monte Carlo