State-dependent vector hybrid linear and nonlinear ARMA modeling: Theory

DOI10.1007/BF01201978MaRDI QIDQ5953488zbMATH OpenFDO

Authors Y. Zheng, Zhiping Lin, D. B. H. Tay

Publication date 24 January 2002

Published in Circuits, Systems, and Signal Processing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/BF01201978

zbMATH Keywords

ARMA modelautoregressive moving average modelexpectation-maximization algorithmfeedforward neural networkhidden Markov chainidentificationmaximum likelihood estimationnonstationary nonlinear processreliable multiple step ahead predictionsecond-order correlation

Mathematics Subject Classification ID

Identification in stochastic control theory (93E12)

Review text

A new model is proposed to represent a general vector nonstationary and nonlinear process by setting up a state-dependent vector hybrid linear and nonlinear autoregressive moving average (SVH-ARMA) model. The linear part of the process is represented by a vector ARMA model, the nonlinear part is represented by a vector nonlinear ARMA model employing a multilayer feedforward neural network, and the nonstationary characteristics are captured with a hidden Markov chain. Based on a unified \(Q\)-likelihood function, an expectation-maximization algorithm for model identification is derived, and the model parameters are estimated by applying a state-dependent training and nonlinear optimization technique iteratively, which finally yields maximum likelihood estimation of model parameters. This model can track the nonstationary variation of a vector linear and/or nonlinear process adaptively and represent a vector linear and/or nonlinear system with low order. Moreover, it is able to characterize and track the long-range, second-order correlation features of many time series and thus can be used for reliable multiple step ahead prediction.

Recommendations

• State-dependent vector hybrid linear and nonlinear ARMA modeling: Applications
• A note on the modelling and analysis of vector ARMA processes with nonstationary innovations
• Current developments in time series modelling
• Modeling nonlinear processes with generalized autoregressions
• Publication:4727239

Cites work

• Analysis of time series subject to changes in regime
• scientific article; zbMATH DE number 3793445 (Why is no real title available?)
• scientific article; zbMATH DE number 51537 (Why is no real title available?)
• scientific article; zbMATH DE number 193741 (Why is no real title available?)
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 3997615 (Why is no real title available?)
• scientific article; zbMATH DE number 777596 (Why is no real title available?)
• On the convergence properties of the EM algorithm
• Optimization. Algorithms and consistent approximations
• State-dependent vector hybrid linear and nonlinear ARMA modeling: Applications

Cited in

(4)

• State representations of ARMA-models
• Hidden multi-resolution ARMA models for nonlinear time series forecasting
• State-dependent vector hybrid linear and nonlinear ARMA modeling: Applications
• LPV-ARX representations of LPV state-space models with affine dependence

This page was built for publication: State-dependent vector hybrid linear and nonlinear ARMA modeling: Theory

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5953488)