Maximum likelihood parameter estimation from incomplete data via the sensitivity equations: the continuous-time case

Author

Charalambous, Charalambos D.

Author_Institution

Dept. of Electr. & Comput. Eng., McGill Univ., Montreal, Que., Canada

Volume

5

fYear

1999

fDate

1999

Firstpage

3412

Abstract

The problem of estimating the parameters for continuous-time partially observed systems is discussed. New exact filters for obtaining maximum likelihood (ML) parameter estimates via the expectation maximization algorithm are derived. The methodology exploits relations between incomplete and complete data likelihood and gradient of likelihood functions, which are derived using Girsanov´s measure transformations. The ML parameter estimates are described by a set of Lyapunov sensitivity equations

Keywords

Kalman filters; Lyapunov methods; continuous time systems; maximum likelihood estimation; sensitivity analysis; stochastic systems; EM algorithm; Girsanov transformation; Kalman filter; Lyapunov method; continuous-time systems; maximum likelihood estimation; parameter estimation; partially observed systems; sensitivity analysis; stochastic systems; Data engineering; Differential equations; Extraterrestrial measurements; Filtering algorithms; Filters; Mathematical model; Maximum likelihood estimation; Parameter estimation; Stochastic systems; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

American Control Conference, 1999. Proceedings of the 1999

Conference_Location

San Diego, CA

ISSN

0743-1619

Print_ISBN

0-7803-4990-3

Type

conf

DOI

10.1109/ACC.1999.782398

Filename

782398