The effect of regularization on variance error
2004 (English)In: IEEE Transactions on Automatic Control, ISSN 0018-9286, E-ISSN 1558-2523, Vol. 49, no 7, 1142-1147 p.Article in journal (Refereed) Published
This note addresses the problem of quantifying the effect of noise induced error(so called variance error) in system estimates found via a regularised cost criterion. It builds on recent work by the authors in which expressions for nonregularised criterions are derived which are exact for finite model order. Those new expressions were established to be very different to previous quantifications that are widely used but based on asymptotic in model order arguments. A key purpose of this note is to expose a rapprochement between these new finite model order, and the preexisting asymptotic model order quantifications. In so doing, a further new result is established. Namely, that variance error in the frequency domain is dependent on the choice of the point about which regularization is affected.
Place, publisher, year, edition, pages
2004. Vol. 49, no 7, 1142-1147 p.
orthonormal bases, parameter estimation, system identification, variance error, asymptotic properties, identification, expressions
Research subject SRA - ICT
IdentifiersURN: urn:nbn:se:kth:diva-23585DOI: 10.1109/tac.2004.831089ISI: 000222760800010ScopusID: 2-s2.0-3843077522OAI: oai:DiVA.org:kth-23585DiVA: diva2:342284
FunderSwedish Research Council, 621-2001-2147
QC 20100525 QC 201201182010-08-102010-08-102012-01-18Bibliographically approved