Path storage in the particle filter

DOI10.1007/S11222-013-9445-XMaRDI QIDQ5962753zbMATH OpenOpenAlexFDO

Authors Pierre E. Jacob, Lawrence M. Murray, Sylvain Rubenthaler

Publication date 23 February 2016

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

Full work available at URL https://arxiv.org/abs/1307.3180

zbMATH Keywords

sequential Monte Carloparticle filterparallel computationmemory cost

Mathematics Subject Classification ID

Numerical analysis or methods applied to Markov chains (65C40) Inference from stochastic processes and prediction (62M20)

Abstract: This article considers the problem of storing the paths generated by a particle filter and more generally by a sequential Monte Carlo algorithm. It provides a theoretical result bounding the expected memory cost by $T+CNlogN$ where $T$ is the time horizon, $N$ is the number of particles and $C$ is a constant, as well as an efficient algorithm to realise this. The theoretical result and the algorithm are illustrated with numerical experiments.

Recommendations

• Particle filters
• Particle Filtering
• On coupling particle filter trajectories
• Particle Filters for Partially Observed Diffusions
• Particle Filtering With Invertible Particle Flow
• Gaussian particle filtering
• Particle Filters for Multiscale Diffusions
• Particle filters for continuous likelihood evaluation and maximisation
• Parallel particle filtering
• Stochastic path summation with memory

Cites work

• scientific article; zbMATH DE number 5919872 (Why is no real title available?)
• scientific article; zbMATH DE number 2051000 (Why is no real title available?)
• scientific article; zbMATH DE number 2106098 (Why is no real title available?)
• Inference in hidden Markov models.
• Long-term stability of sequential Monte Carlo methods under verifiable conditions
• On disturbance state-space models and the particle marginal Metropolis-Hastings sampler
• Particle Gibbs with ancestor sampling
• Particle approximations of the score and observed information matrix in state space models with application to parameter estimation
• Sequential Monte Carlo Methods for Dynamic Systems
• Sequential Monte Carlo Methods in Practice
• Stability properties of some particle filters
• The convergence to equilibrium of neutral genetic models
• The time back to the most recent common ancestor in exchangeable population models
• Uniform time average consistency of Monte Carlo particle filters
• \(\mathrm{SMC}^2\): an efficient algorithm for sequential analysis of state space models

Cited in

(14)

• Particle-based online estimation of tangent filters with application to parameter estimation in nonlinear state-space models
• Simple conditions for convergence of sequential Monte Carlo genealogies with applications
• Numerically stable online estimation of variance in particle filters
• A lagged particle filter for stable filtering of certain high-dimensional state-space models
• Asymptotic genealogies of interacting particle systems with an application to sequential Monte Carlo
• Sequential sampling of junction trees for decomposable graphs
• On the two-filter approximations of marginal smoothing distributions in general state-space models
• Smoothing with couplings of conditional particle filters
• Biased online parameter inference for state-space models
• On particle Gibbs sampling
• Weak convergence of non-neutral genealogies to Kingman's coalescent
• Automatically adapting the number of state particles in \(\text{SMC}^2\)
• Inferring medication adherence from time-varying health measures
• Controlled sequential Monte Carlo

Describes a project that uses

Uses Software

• LibBi

This page was built for publication: Path storage in the particle filter

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5962753)