Testing for shifts in trend with an integrated or stationary noise component

Pierre Perron, Tomoyoshi Yabu

Research output: Contribution to journalArticle

124 Citations (Scopus)

Abstract

We consider testing for structural changes in the trend function of a time series without any prior knowledge of whether the noise component is stationary or integrated. Following Perron and Yabu (2009), we consider a quasi-feasible generalized least squares procedure that uses a super-efficient estimate of the sum of the autoregressive parameters α when α = 1. This allows tests of basically the same size with stationary or integrated noise regardless of whether the break is known or unknown, provided that the Exp functional of Andrews and Ploberger (1994) is used in the latter case. To improve the finite-sample properties, we use the bias-corrected version of the estimate of α proposed by Roy and Fuller (2001). Our procedure has a power function close to that attainable if we knew the true value of α in many cases. We also discuss the extension to the case of multiple breaks.

Original languageEnglish
Pages (from-to)369-396
Number of pages28
JournalJournal of Business and Economic Statistics
Volume27
Issue number3
DOIs
Publication statusPublished - 2009

Fingerprint

Testing
Generalized Least Squares
Structural Change
Power Function
trend
Prior Knowledge
structural change
Estimate
time series
Time series
Unknown
knowledge
Values
Trends
Integrated
Prior knowledge
Structural change
Finite sample properties
Multiple breaks
Generalized least squares

Keywords

  • Generalized least squares procedure
  • Median-unbiased estimate
  • Structural change
  • Super-efficient estimate
  • Unit root

ASJC Scopus subject areas

  • Statistics and Probability
  • Economics and Econometrics
  • Social Sciences (miscellaneous)
  • Statistics, Probability and Uncertainty

Cite this

Testing for shifts in trend with an integrated or stationary noise component. / Perron, Pierre; Yabu, Tomoyoshi.

In: Journal of Business and Economic Statistics, Vol. 27, No. 3, 2009, p. 369-396.

Research output: Contribution to journalArticle

@article{0f5bc52dcf584a668f58736079a2bfc7,
title = "Testing for shifts in trend with an integrated or stationary noise component",
abstract = "We consider testing for structural changes in the trend function of a time series without any prior knowledge of whether the noise component is stationary or integrated. Following Perron and Yabu (2009), we consider a quasi-feasible generalized least squares procedure that uses a super-efficient estimate of the sum of the autoregressive parameters α when α = 1. This allows tests of basically the same size with stationary or integrated noise regardless of whether the break is known or unknown, provided that the Exp functional of Andrews and Ploberger (1994) is used in the latter case. To improve the finite-sample properties, we use the bias-corrected version of the estimate of α proposed by Roy and Fuller (2001). Our procedure has a power function close to that attainable if we knew the true value of α in many cases. We also discuss the extension to the case of multiple breaks.",
keywords = "Generalized least squares procedure, Median-unbiased estimate, Structural change, Super-efficient estimate, Unit root",
author = "Pierre Perron and Tomoyoshi Yabu",
year = "2009",
doi = "10.1198/jbes.2009.07268",
language = "English",
volume = "27",
pages = "369--396",
journal = "Journal of Business and Economic Statistics",
issn = "0735-0015",
publisher = "American Statistical Association",
number = "3",

}

TY - JOUR

T1 - Testing for shifts in trend with an integrated or stationary noise component

AU - Perron, Pierre

AU - Yabu, Tomoyoshi

PY - 2009

Y1 - 2009

N2 - We consider testing for structural changes in the trend function of a time series without any prior knowledge of whether the noise component is stationary or integrated. Following Perron and Yabu (2009), we consider a quasi-feasible generalized least squares procedure that uses a super-efficient estimate of the sum of the autoregressive parameters α when α = 1. This allows tests of basically the same size with stationary or integrated noise regardless of whether the break is known or unknown, provided that the Exp functional of Andrews and Ploberger (1994) is used in the latter case. To improve the finite-sample properties, we use the bias-corrected version of the estimate of α proposed by Roy and Fuller (2001). Our procedure has a power function close to that attainable if we knew the true value of α in many cases. We also discuss the extension to the case of multiple breaks.

AB - We consider testing for structural changes in the trend function of a time series without any prior knowledge of whether the noise component is stationary or integrated. Following Perron and Yabu (2009), we consider a quasi-feasible generalized least squares procedure that uses a super-efficient estimate of the sum of the autoregressive parameters α when α = 1. This allows tests of basically the same size with stationary or integrated noise regardless of whether the break is known or unknown, provided that the Exp functional of Andrews and Ploberger (1994) is used in the latter case. To improve the finite-sample properties, we use the bias-corrected version of the estimate of α proposed by Roy and Fuller (2001). Our procedure has a power function close to that attainable if we knew the true value of α in many cases. We also discuss the extension to the case of multiple breaks.

KW - Generalized least squares procedure

KW - Median-unbiased estimate

KW - Structural change

KW - Super-efficient estimate

KW - Unit root

UR - http://www.scopus.com/inward/record.url?scp=74049143004&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74049143004&partnerID=8YFLogxK

U2 - 10.1198/jbes.2009.07268

DO - 10.1198/jbes.2009.07268

M3 - Article

AN - SCOPUS:74049143004

VL - 27

SP - 369

EP - 396

JO - Journal of Business and Economic Statistics

JF - Journal of Business and Economic Statistics

SN - 0735-0015

IS - 3

ER -