Particle-based, online estimation of tangent filters with application to parameter estimation in nonlinear state-space models
Preprint
Abstract
This paper presents a novel algorithm for efficient online estimation of the filter derivatives in general hidden Markov models. The algorithm, which has a linear computational complexity and very limited memory requirements, is furnished with a number of convergence results, including a central limit theorem with an asymptotic variance that can be shown to be uniformly bounded in time. Using the proposed filter derivative estimator we design a recursive maximum likelihood algorithm updating the parameters according the gradient of the one-step predictor log-likelihood. The efficiency of this online parameter estimation scheme is illustrated in a simulation study.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- ON THE TWO-FILTER APPROXIMATIONS OF MARGINAL SMOOTHING DISTRIBUTIONS IN GENERAL STATE-SPACE MODELSAdvances in Applied Probability, 2018
- Path storage in the particle filterStatistics and Computing, 2013
- Particle approximations of the score and observed information matrix in state space models with application to parameter estimationBiometrika, 2011
- Analyticity, Convergence, and Convergence Rate of Recursive Maximum-Likelihood Estimation in Hidden Markov ModelsIEEE Transactions on Information Theory, 2010
- A backward particle interpretation of Feynman-Kac formulaeESAIM: Mathematical Modelling and Numerical Analysis, 2010
- A sequential smoothing algorithm with linear computational costBiometrika, 2010
- Parameter estimation in general state-space models using particle methodsAnnals of the Institute of Statistical Mathematics, 2003
- On sequential Monte Carlo sampling methods for Bayesian filteringStatistics and Computing, 2000
- Exponential forgetting and geometric ergodicity in hidden Markov models1997
- General results on the convergence of stochastic algorithmsIEEE Transactions on Automatic Control, 1996