A self-organizing state space model and simplex initial distribution search

DOI10.1007/S00180-007-0027-2MaRDI QIDQ626199zbMATH OpenOpenAlexFDO

Authors Koiti Yano

Publication date 22 February 2011

Published in Computational Statistics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s00180-007-0027-2

zbMATH Keywords

maximum likelihoodMonte Carlo particle filteringNelder Mead algorithm

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Inference from stochastic processes and prediction (62M20) Monte Carlo methods (65C05) Filtering in stochastic control theory (93E11)

Review text

A Monte Carlo particle (MCP) filter is considered for nonlinear non-Gaussian state-space (SS) models. A self-organizing SS (SOSS) is discussed where the unknown parameters of the model are included in the augmented state vector. A version of the simplex Nelder-Mead maximization algorithm is proposed for maximum likelihood estimation calculation in SOSS with MCP filter. Numerical examples are presented for linear and nonlinear Gaussian and non-Gaussian SS, a stochastic volatility model and a threshold autoregressive model.

Recommendations

• Nonlinear and non-Gaussian state-space modeling with Monte Carlo simulations
• Monte Carlo maximum likelihood estimation for non-Gaussian state space models
• Sequential Monte Carlo smoothing with application to parameter estimation in nonlinear state space models
• Estimation of unknown parameters in nonlinear and non-Gaussian state-space models
• Parameter estimation in general state-space models using particle methods

Cites work

• A Simplex Method for Function Minimization
• Dynamic Generalized Linear Models and Bayesian Forecasting
• scientific article; zbMATH DE number 1666093 (Why is no real title available?)
• scientific article; zbMATH DE number 48093 (Why is no real title available?)
• Non-Gaussian State-Space Modeling of Nonstationary Time Series
• Nonlinear Bayesian estimation using Gaussian sum approximations
• Parameter estimation in general state-space models using particle methods
• Sequential Monte Carlo Methods in Practice
• Stochastic processes and filtering theory

Cited in

(1)

• A new method for sequential learning of states and parameters for state-space models: the particle swarm learning optimization

This page was built for publication: A self-organizing state space model and simplex initial distribution search

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q626199)