Estimation of the autoregressive operator by wavelet packets

In this paper, we suggest to use wavelet packet bases as an alternative to the widely used principal component analysis in an estimation of the functional autoregressive processes. By extending the notion of the so-called non-standard form of operators representation, we search for the “best” basis...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Statistics & probability letters 2009-01, Vol.79 (1), p.38-43
Hauptverfasser:	Laukaitis, Algirdas, Vasilecas, Olegas, Laukaitis, Ricardas
Format:	Artikel
Sprache:	eng
Schlagworte:	Exact sciences and technology General topics Mathematics Multivariate analysis Probability and statistics Probability theory and stochastic processes Sciences and techniques of general use Statistics
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	In this paper, we suggest to use wavelet packet bases as an alternative to the widely used principal component analysis in an estimation of the functional autoregressive processes. By extending the notion of the so-called non-standard form of operators representation, we search for the “best” basis on a criterion of highest correlation between pairs of wavelet packet coefficients.
ISSN:	0167-7152 1879-2103
DOI:	10.1016/j.spl.2008.07.011