Finite dimensional filters for random parameter AR models

Author

Evans, Jamie ; Krishnamurthy, Vikram

Author_Institution

Dept. of Electr. & Electron. Eng., Melbourne Univ., Parkville, Vic., Australia

Volume

5

fYear

1997

fDate

4-6 Jun 1997

Firstpage

2836

Abstract

In this paper exact finite dimensional filters are derived for a class of doubly stochastic autoregressive models. The parameters of the doubly stochastic autoregressive process vary according to a nonlinear function of a Gauss-Markov process. We develop a difference equation for the evolution of an unnormalized conditional density related to the state of the doubly stochastic autoregressive process. We then give a characterization of the general solution followed by examples for which the state of the filter is determined by a finite number of sufficient statistics. These new finite dimensional filters are built upon the discrete-time Kalman filter

Keywords

Kalman filters; autoregressive processes; difference equations; filtering theory; probability; stochastic processes; Gauss-Markov process; autoregressive models; difference equation; discrete-time Kalman filter; finite dimensional filters; probability; stochastic AR models; unnormalized conditional density; Covariance matrix; Difference equations; Filters; Gaussian processes; Hidden Markov models; Random processes; Signal processing; Statistics; Stochastic processes; Stochastic resonance;

fLanguage

English

Publisher

ieee

Conference_Titel

American Control Conference, 1997. Proceedings of the 1997

Conference_Location

Albuquerque, NM

ISSN

0743-1619

Print_ISBN

0-7803-3832-4

Type

conf

DOI

10.1109/ACC.1997.611973

Filename

611973