Log-periodogram estimation of the memory parameter of a long-memory process under trend

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

  • Philipp Sibbertsen

Externe Organisationen

  • Technische Universität Dortmund
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)261-268
Seitenumfang8
FachzeitschriftStatistics and Probability Letters
Jahrgang61
Ausgabenummer3
Frühes Online-Datum1 Nov. 2002
PublikationsstatusVeröffentlicht - 1 Feb. 2003
Extern publiziertJa

Abstract

We show that log-periodogram-based estimators for the memory parameter in a stationary invertible long memory process do not confuse small trends withlong-range dependence. In the case of slowly decaying trends we show by Monte Carlo methods that the tapered periodogram is quite robust against these trends and reduces the bias obtained when employing the standard log-periodogram estimator. Thus, comparing the tapered and the non-tapered estimator gives a tool at hand for distinguishing slowly decaying trends and long-range dependence.

ASJC Scopus Sachgebiete

Zitieren

Log-periodogram estimation of the memory parameter of a long-memory process under trend. / Sibbertsen, Philipp.
in: Statistics and Probability Letters, Jahrgang 61, Nr. 3, 01.02.2003, S. 261-268.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Sibbertsen P. Log-periodogram estimation of the memory parameter of a long-memory process under trend. Statistics and Probability Letters. 2003 Feb 1;61(3):261-268. Epub 2002 Nov 1. doi: 10.1016/S0167-7152(02)00358-9
Download
@article{139f6e46131b4968b5fbf3d05152c220,
title = "Log-periodogram estimation of the memory parameter of a long-memory process under trend",
abstract = "We show that log-periodogram-based estimators for the memory parameter in a stationary invertible long memory process do not confuse small trends withlong-range dependence. In the case of slowly decaying trends we show by Monte Carlo methods that the tapered periodogram is quite robust against these trends and reduces the bias obtained when employing the standard log-periodogram estimator. Thus, comparing the tapered and the non-tapered estimator gives a tool at hand for distinguishing slowly decaying trends and long-range dependence.",
keywords = "Log-periodogram regression, Long memory, Trends",
author = "Philipp Sibbertsen",
note = "Funding Information: The computational assistance of Klaus Nordhausen and the helpful comments of the unknown referee are gratefully acknowledged. The author thanks the Volkswagenstiftung for financial support. ",
year = "2003",
month = feb,
day = "1",
doi = "10.1016/S0167-7152(02)00358-9",
language = "English",
volume = "61",
pages = "261--268",
journal = "Statistics and Probability Letters",
issn = "0167-7152",
publisher = "Elsevier",
number = "3",

}

Download

TY - JOUR

T1 - Log-periodogram estimation of the memory parameter of a long-memory process under trend

AU - Sibbertsen, Philipp

N1 - Funding Information: The computational assistance of Klaus Nordhausen and the helpful comments of the unknown referee are gratefully acknowledged. The author thanks the Volkswagenstiftung for financial support.

PY - 2003/2/1

Y1 - 2003/2/1

N2 - We show that log-periodogram-based estimators for the memory parameter in a stationary invertible long memory process do not confuse small trends withlong-range dependence. In the case of slowly decaying trends we show by Monte Carlo methods that the tapered periodogram is quite robust against these trends and reduces the bias obtained when employing the standard log-periodogram estimator. Thus, comparing the tapered and the non-tapered estimator gives a tool at hand for distinguishing slowly decaying trends and long-range dependence.

AB - We show that log-periodogram-based estimators for the memory parameter in a stationary invertible long memory process do not confuse small trends withlong-range dependence. In the case of slowly decaying trends we show by Monte Carlo methods that the tapered periodogram is quite robust against these trends and reduces the bias obtained when employing the standard log-periodogram estimator. Thus, comparing the tapered and the non-tapered estimator gives a tool at hand for distinguishing slowly decaying trends and long-range dependence.

KW - Log-periodogram regression

KW - Long memory

KW - Trends

UR - http://www.scopus.com/inward/record.url?scp=0037307569&partnerID=8YFLogxK

U2 - 10.1016/S0167-7152(02)00358-9

DO - 10.1016/S0167-7152(02)00358-9

M3 - Article

AN - SCOPUS:0037307569

VL - 61

SP - 261

EP - 268

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

SN - 0167-7152

IS - 3

ER -