Adaptive non-linear time-series estimation based on hidden Markov models

Author

Krishnamurthy, Vikram

Author_Institution

Dept. of Syst. Eng., Australian Nat. Univ., Canberra, ACT

fYear

1993

fDate

15-17 Dec 1993

Firstpage

720

Abstract

In this paper we propose maximum-likelihood (ML) estimation schemes for the parameters and states of ARMAX systems when the input is a finite-state Markov chain. Such models have applications in econometrics, speech processing, communication systems and neuro-biological signal processing. We derive the ML model estimates using the expectation maximization (EM) algorithm. We then develop two sequential or “online” estimation schemes: Recursive EM algorithm and a gradient based scheme

Keywords

hidden Markov models; maximum likelihood estimation; nonlinear systems; parameter estimation; state estimation; time series; ARMAX systems; ML model estimates; adaptive nonlinear time-series estimation; finite-state Markov chain; gradient based scheme; hidden Markov models; maximum-likelihood estimation schemes; online estimation; parameter estimation; recursive EM algorithm; recursive expectation maximization algorithm; sequential estimation; state estimation; Convergence; Delay; Econometrics; Equations; Hidden Markov models; Maximum likelihood estimation; Recursive estimation; Speech; State estimation; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 1993., Proceedings of the 32nd IEEE Conference on

Conference_Location

San Antonio, TX

Print_ISBN

0-7803-1298-8

Type

conf

DOI

10.1109/CDC.1993.325055

Filename

325055