Details
| Original language | English |
|---|---|
| Pages (from-to) | 357-370 |
| Number of pages | 14 |
| Journal | Statistical papers |
| Volume | 53 |
| Issue number | 2 |
| Early online date | 26 Jun 2010 |
| Publication status | Published - May 2012 |
Abstract
We show that the CUSUM-squared based test for a change in persistence by Leybourne et al. (J Time Ser Anal 28:408-433, 2007) is not robust against shifts in the mean. A mean shift leads to serious size distortions. Therefore, adjusted critical values are needed when it is known that the data generating process has a mean shift. These are given for the case of one mean break. Response curves for the critical values are derived and a Monte Carlo study showing the size and power properties under this general de-trending is given.
Keywords
- Break in persistence, Level shift, Long memory, Structural break
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Decision Sciences(all)
- Statistics, Probability and Uncertainty
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Statistical papers, Vol. 53, No. 2, 05.2012, p. 357-370.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Testing for a break in persistence under long-range dependencies and mean shifts
AU - Sibbertsen, Philipp
AU - Willert, Juliane
N1 - Funding Information: Acknowledgments We would like to thank Robinson Kruse, an anonymous referee and the participants of the Pfingsttagung 2009 of the German Statistical Society in Merseburg for their helpful comments and suggestions. The financial support of the Fritz Thyssen Stiftung für Wissenschaftsfoerderung is gratefully acknowledged.
PY - 2012/5
Y1 - 2012/5
N2 - We show that the CUSUM-squared based test for a change in persistence by Leybourne et al. (J Time Ser Anal 28:408-433, 2007) is not robust against shifts in the mean. A mean shift leads to serious size distortions. Therefore, adjusted critical values are needed when it is known that the data generating process has a mean shift. These are given for the case of one mean break. Response curves for the critical values are derived and a Monte Carlo study showing the size and power properties under this general de-trending is given.
AB - We show that the CUSUM-squared based test for a change in persistence by Leybourne et al. (J Time Ser Anal 28:408-433, 2007) is not robust against shifts in the mean. A mean shift leads to serious size distortions. Therefore, adjusted critical values are needed when it is known that the data generating process has a mean shift. These are given for the case of one mean break. Response curves for the critical values are derived and a Monte Carlo study showing the size and power properties under this general de-trending is given.
KW - Break in persistence
KW - Level shift
KW - Long memory
KW - Structural break
UR - http://www.scopus.com/inward/record.url?scp=84860176714&partnerID=8YFLogxK
U2 - 10.1007/s00362-010-0342-5
DO - 10.1007/s00362-010-0342-5
M3 - Article
AN - SCOPUS:84860176714
VL - 53
SP - 357
EP - 370
JO - Statistical papers
JF - Statistical papers
SN - 0932-5026
IS - 2
ER -