Nonlinear filters based on taylor series expansions∗

DOI10.1080/03610929608831763MaRDI QIDQ4337190zbMATH OpenOpenAlexFDO

Authors Hisashi Tanizaki, Roberto S. Mariano

Publication date 19 May 1997

Published in Communications in Statistics: Theory and Methods (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/03610929608831763, https://ink.library.smu.edu.sg/soe_research/302

zbMATH Keywords

extended Kalman filterMonte-Carlo simulation filterrandom drawssecond-order nonlinear filterTaylor series expansion

Mathematics Subject Classification ID

Inference from stochastic processes and prediction (62M20)

Review text

The nonlinear filters based on Taylor series approximation are broadly used for computational simplicity, even though their filtering estimates are clearly biased. In this paper, first, we analyze what is approximated when we apply the expanded nonlinear functions to the standard linear recursive Kalman filter algorithm. Next, since the state variables \(\alpha_t\) and \(\alpha_{t-1}\), are approximated as a conditional normal distribution given information up to time \(t-1\) (i.e., \(I_{t-1}\)) in approximation of the Taylor series expansion, it might be appropriate to evaluate each expectation by generating normal random numbers of \(\alpha_t\) and \(\alpha_{t-1}\) given \(I_{t-1}\), und those of the error terms \(\varepsilon_t\) and \(\eta_t\). Thus, we propose the Monte-Carlo simulation filter using normal random draws. Finally we perform two Monte-Carlo experiments, where we obtain the result that the Monte-Carlo simulation filter has a superior performance over the nonlinear filters such as the extended Kalman filter and the second-order nonlinear filter.

Recommendations

• scientific article; zbMATH DE number 1053273
• Nonlinear and nonnormal filters using Monte Carlo methods
• Nonlinear filters. Estimation and applications
• Nonlinear and non-Gaussian state-space modeling with Monte Carlo simulations

Cites work

• A comparison of three non-linear filters
• Asymptotic Behavior of Predictors in a Nonlinear Simultaneous System
• Asymptotically exact confidence intervals of cusum and cusumsq tests: a numerical derivation using simulation technique
• scientific article; zbMATH DE number 3727458 (Why is no real title available?)
• Measures of Deterministic Prediction Bias in Nonlinear Models
• Non-Gaussian State-Space Modeling of Nonstationary Time Series
• Nonlinear Bayesian estimation using Gaussian sum approximations
• Recursive Bayesian estimation using Gaussian sums
• Recursive Bayesian estimation using piecewise constant approximations
• Residual-Based Procedures for Prediction and Estimation in a Nonlinear Simultaneous System

Cited in

(9)

• Nonlinear filters. Estimation and applications
• Nonlinear and non-Gaussian state-space modeling with Monte Carlo simulations
• Nonlinear and nonnormal filters using Monte Carlo methods
• scientific article; zbMATH DE number 1053273 (Why is no real title available?)
• On markov chain monte carlo methods for nonlinear and non-gaussian state-space models
• Nonlinear and non-gaussian state estimation: A quasi-optimal estimator
• Modal Kalman filter
• Bayesian estimation of state-space models using the Metropolis-Hastings algorithm within Gibbs sampling.
• Nonlinear filters approximations by cumulant function expansion: The polynomial case

This page was built for publication: Nonlinear filters based on taylor series expansions^∗

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4337190)