Stability properties of some particle filters
arXiv:1109.6779 · doi:10.1214/12-AAP909
Abstract
Under multiplicative drift and other regularity conditions, it is established that the asymptotic variance associated with a particle filter approximation of the prediction filter is bounded uniformly in time, and the nonasymptotic, relative variance associated with a particle approximation of the normalizing constant is bounded linearly in time. The conditions are demonstrated to hold for some hidden Markov models on noncompact state spaces. The particle stability results are obtained by proving -norm multiplicative stability and exponential moment results for the underlying Feynman-Kac formulas.
Published in at http://dx.doi.org/10.1214/12-AAP909 the Annals of Applied Probability (http://www.imstat.org/aap/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (6)
- Recursive Monte Carlo filters: Algorithms and theoretical analysis
- On adaptive resampling strategies for sequential Monte Carlo methods
- Sequential Monte Carlo smoothing for general state space hidden Markov models
- Uniform Time Average Consistency of Monte Carlo Particle Filters
- Linear Variance Bounds for Particle Approximations of Time-Homogeneous Feynman-Kac Formulae
- Tree based functional expansions for Feynman--Kac particle models
Cited by in corpus (11)
- On Particle Methods for Parameter Estimation in State-Space Models
- The Optimal Arbitrary-Proportional Finite-Set-Partitioning
- On the role of interaction in sequential Monte Carlo algorithms
- Bayesian model comparison with un-normalised likelihoods
- Importance sampling type estimators based on approximate marginal MCMC
- Twisted particle filters
- On the Mathematical Theory of Ensemble (Linear-Gaussian) Kalman-Bucy Filtering
- Approximate Smoothing and Parameter Estimation in High-Dimensional State-Space Models
- Limit theorems for sequential MCMC methods
- Central Limit Theorems for Coupled Particle Filters
- Probing robustness of nonlinear filter stability numerically using Sinkhorn divergence