Reduced-rank linear regression

Author

Stoica, P. ; Viberg, M.

Author_Institution

Syst. & Control Group, Uppsala Univ., Sweden

fYear

1996

fDate

24-26 Jun 1996

Firstpage

542

Lastpage

545

Abstract

This paper considers the problem of maximum likelihood (ML) estimation for reduced-rank linear regression equations with noise of arbitrary covariance. An explicit expression for the ML estimate of the regression matrix is derived. A generalized likelihood ratio (GLRT) test as also proposed, for estimating the rank of the regression matrix. Computer simulations and numerical examples indicate the superiority of the proposed estimator, as compared to a traditional least-squares approach that does not exploit the reduced rank property in an optimal way

Keywords

correlation methods; matrix algebra; maximum likelihood estimation; GLRT test; ML estimate; computer simulations; covariance; generalized likelihood ratio test; maximum likelihood estimation; reduced-rank linear regression; reduced-rank linear regression equations; regression matrix; truncated canonical correlation decomposition; Array signal processing; Control systems; Covariance matrix; Equations; Linear regression; Maximum likelihood estimation; Parameter estimation; Sensor arrays; State estimation; White noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Statistical Signal and Array Processing, 1996. Proceedings., 8th IEEE Signal Processing Workshop on (Cat. No.96TB10004

Conference_Location

Corfu

Print_ISBN

0-8186-7576-4

Type

conf

DOI

10.1109/SSAP.1996.534934

Filename

534934