Particle Metropolis-Hastings using gradient and Hessian information

DOI10.1007/S11222-014-9510-0zbMATH Open1331.62134arXiv1311.0686OpenAlexW3099693596MaRDI QIDQ5963543FDOQ5963543

Johan Dahlin, Fredrik Lindsten, Thomas B. Schön

Publication date: 22 February 2016

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: Particle Metropolis-Hastings (PMH) allows for Bayesian parameter inference in nonlinear state space models by combining Markov chain Monte Carlo (MCMC) and particle filtering. The latter is used to estimate the intractable likelihood. In its original formulation, PMH makes use of a marginal MCMC proposal for the parameters, typically a Gaussian random walk. However, this can lead to a poor exploration of the parameter space and an inefficient use of the generated particles. We propose a number of alternative versions of PMH that incorporate gradient and Hessian information about the posterior into the proposal. This information is more or less obtained as a byproduct of the likelihood estimation. Indeed, we show how to estimate the required information using a fixed-lag particle smoother, with a computational cost growing linearly in the number of particles. We conclude that the proposed methods can: (i) decrease the length of the burn-in phase, (ii) increase the mixing of the Markov chain at the stationary phase, and (iii) make the proposal distribution scale invariant which simplifies tuning.

Full work available at URL: https://arxiv.org/abs/1311.0686

zbMATH Keywords

sequential Monte Carloparameter inferenceparticle Markov chain Monte Carlofixed-lag particle smoothingmanifold MALA

Mathematics Subject Classification ID

Bayesian inference (62F15) Numerical analysis or methods applied to Markov chains (65C40) Information theory (general) (94A15)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• The pseudo-marginal approach for efficient Monte Carlo computations
• Analysis of a nonreversible Markov chain sampler.
• Inference in hidden Markov models.
• Filtering via Simulation: Auxiliary Particle Filters
• Particle Markov Chain Monte Carlo Methods
• Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator
• Particle approximations of the score and observed information matrix in state space models with application to parameter estimation
• Langevin diffusions and Metropolis-Hastings algorithms
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Riemann Manifold Langevin and Hamiltonian Monte Carlo Methods
• Sequential Monte Carlo smoothing with application to parameter estimation in nonlinear state space models
• On the efficiency of pseudo-marginal random walk Metropolis algorithms
• BAYESIAN INFERENCE BASED ONLY ON SIMULATED LIKELIHOOD: PARTICLE FILTER ANALYSIS OF DYNAMIC ECONOMIC MODELS
• On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
• Particle Metropolis-Hastings using gradient and Hessian information
• Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
• Particle Metropolis-adjusted Langevin algorithms
• Some applications of nonlinear and non-Gaussian state–space modelling by means of hidden Markov models

Cited In (9)

• On particle methods for parameter estimation in state-space models
• Regularized zero-variance control variates
• Augmentation schemes for particle MCMC
• Particle Metropolis-Hastings using gradient and Hessian information
• A rare event approach to high-dimensional approximate Bayesian computation
• A flexible particle Markov chain Monte Carlo method
• Filtering and estimation for a class of stochastic volatility models with intractable likelihoods
• A second-order iterated smoothing algorithm
• Introduction to ``Particle Metropolis-Hastings using gradient and Hessian information by J. Dahlin, F. Lindsten, T. Schön

This page was built for publication: Particle Metropolis-Hastings using gradient and Hessian information

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5963543)