On the asymptotic accuracy of pseudo-linear regression algorithms

The accuracy properties of a general pseudo-linear regression (PLR) method are examined. Both off-line and on-line algorithms are considered. Assuming the parameter estimates converge they will under weak conditions be asymptotically gaussian distributed. Expressions for the corresponding covariance...

Full description

Saved in:
Bibliographic Details
Published in:International journal of control Vol. 39; no. 1; pp. 115 - 126
Main Authors: STOICA, P, SODERSTROM, T, AHLEN, A, SOLBRAND, G
Format: Journal Article
Language:English
Published: London Taylor & Francis Group 01.01.1984
Taylor & Francis
Subjects:
Applied sciences
Computer science; control theory; systems
Control theory. Systems
Exact sciences and technology
Modelling and identification
Parameter estimation
System identification
Regression model
Algorithm
ISSN:0020-7179, 1366-5820
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The accuracy properties of a general pseudo-linear regression (PLR) method are examined. Both off-line and on-line algorithms are considered. Assuming the parameter estimates converge they will under weak conditions be asymptotically gaussian distributed. Expressions for the corresponding covariance matrices are given. It is shown that the asymptotic covariance matrix of the off-line PLR algorithm is bounded from above by the matrix corresponding to the on-line PLR algorithm and from below by that corresponding to the prediction error method. Some simple numerical illustrations of the theoretical results are also included.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0020-7179
1366-5820
DOI:10.1080/00207178408933152