Nested particle filters for online parameter estimation in discrete-time state-space Markov models

DOI10.3150/17-BEJ954MaRDI QIDQ1708992zbMATHOpenAlexFDO

Authors Dan Crisan, Joaquín Míguez

Publication date 27 March 2018

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1308.1883

zbMATH Keywords

parameter estimationMonte Carloerror boundsstate space modelsparticle filteringrecursive algorithmsmodel inference

Mathematics Subject Classification ID

Bayesian inference (62F15) Markov processes: estimation; hidden Markov models (62M05) Inference from stochastic processes and prediction (62M20) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10)

Abstract: We address the problem of approximating the posterior probability distribution of the fixed parameters of a state-space dynamical system using a sequential Monte Carlo method. The proposed approach relies on a nested structure that employs two layers of particle filters to approximate the posterior probability measure of the static parameters and the dynamic state variables of the system of interest, in a vein similar to the recent "sequential Monte Carlo square" (SMC${}^2$) algorithm. However, unlike the SMC${}^2$ scheme, the proposed technique operates in a purely recursive manner. In particular, the computational complexity of the recursive steps of the method introduced herein is constant over time. We analyse the approximation of integrals of real bounded functions with respect to the posterior distribution of the system parameters computed via the proposed scheme. As a result, we prove, under regularity assumptions, that the approximation errors vanish asymptotically in $L_p$ ($pge1$) with convergence rate proportional to $frac1sqrtN+frac1sqrtM$, where $N$ is the number of Monte Carlo samples in the parameter space and $NimesM$ is the number of samples in the state space. This result also holds for the approximation of the joint posterior distribution of the parameters and the state variables. We discuss the relationship between the SMC${}^2$ algorithm and the new recursive method and present a simple example in order to illustrate some of the theoretical findings with computer simulations.

Recommendations

• Uniform convergence over time of a nested particle filtering scheme for recursive parameter estimation in state-space Markov models
• A Kalman particle filter for online parameter estimation with applications to affine models
• Particle-based online estimation of tangent filters with application to parameter estimation in nonlinear state-space models
• Parameter estimation in general state-space models using particle methods
• \(\mathrm{SMC}^2\): an efficient algorithm for sequential analysis of state space models

Cites work

• scientific article; zbMATH DE number 1666084 (Why is no real title available?)
• scientific article; zbMATH DE number 1666085 (Why is no real title available?)
• scientific article; zbMATH DE number 1666093 (Why is no real title available?)
• scientific article; zbMATH DE number 2061746 (Why is no real title available?)
• scientific article; zbMATH DE number 1500585 (Why is no real title available?)
• scientific article; zbMATH DE number 2106098 (Why is no real title available?)
• A Particle Filtering Scheme for Processing Time Series Corrupted by Outliers
• A population Monte Carlo scheme with transformed weights and its application to stochastic kinetic models
• A sequential particle filter method for static models
• A survey of convergence results on particle filtering methods for practitioners
• Adaptive joint detection and decoding in flat-fading channels via mixture Kalman filtering
• Deterministic Nonperiodic Flow
• Dimensional reduction for a Bayesian filter
• Following a moving target -- Monte Carlo inference for dynamic Bayesian models
• On particle methods for parameter estimation in state-space models
• On the convergence of two sequential Monte Carlo methods for maximum a posteriori sequence estimation and stochastic global optimization
• On the stability of sequential Monte Carlo methods in high dimensions
• Parameter estimation and asymptotic stability in stochastic filtering
• Particle Markov chain Monte Carlo for efficient numerical simulation
• Particle approximations of the score and observed information matrix in state space models with application to parameter estimation
• Particle learning and smoothing
• Sequential Imputations and Bayesian Missing Data Problems
• Sequential Monte Carlo Methods for Dynamic Systems
• Sequential Monte Carlo methods for nonlinear discrete-time filtering
• Sequential Monte Carlo smoothing with application to parameter estimation in nonlinear state space models
• Uniform Stability of a Particle Approximation of the Optimal Filter Derivative
• \(\mathrm{SMC}^2\): an efficient algorithm for sequential analysis of state space models

Cited in

(22)

• Nudging the particle filter
• Particle-based online estimation of tangent filters with application to parameter estimation in nonlinear state-space models
• Inference via low-dimensional couplings
• Parallelizing particle filters with butterfly interactions
• Sequential estimation of temporally evolving latent space network models
• Parallel sequential Monte Carlo for stochastic gradient-free nonconvex optimization
• Sequential Bayesian inference for static parameters in dynamic state space models
• A sequential importance sampling filter with a new proposal distribution for state and parameter estimation of nonlinear dynamical systems
• On the performance of particle filters with adaptive number of particles
• On-Line Inference for Hidden Markov Models via Particle Filters
• On classical and Bayesian asymptotics in state space stochastic differential equations
• Optimal estimation of amplitude and phase modulated signals
• Boolean Kalman filter and smoother under model uncertainty
• Uniform convergence over time of a nested particle filtering scheme for recursive parameter estimation in state-space Markov models
• Biased online parameter inference for state-space models
• A rare event approach to high-dimensional approximate Bayesian computation
• Convergence of Regularized Particle Filters for Stochastic Reaction Networks
• A particle filter for stochastic advection by Lie transport: a case study for the damped and forced incompressible two-dimensional Euler equation
• Automatically adapting the number of state particles in \(\text{SMC}^2\)
• Stochastic filtering methods in electronic trading
• A Kalman particle filter for online parameter estimation with applications to affine models
• Particle approximations of the score and observed information matrix in state space models with application to parameter estimation

This page was built for publication: Nested particle filters for online parameter estimation in discrete-time state-space Markov models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1708992)