Incidental trends and the power of panel unit root tests

Hyungsik Roger Moon, Benoit Perron, Peter C.B. Phillips

Research output: Contribution to journalArticle

54 Citations (Scopus)

Abstract

The asymptotic local power of various panel unit root tests is investigated. The (Gaussian) power envelope is obtained under homogeneous and heterogeneous alternatives. The envelope is compared with the asymptotic power functions for the pooled t-test, the Ploberger and Phillips [2002. Optimal testing for unit roots in panel data. Mimeo] test, and a point optimal test in neighborhoods of unity that are of order n- 1 / 4 T- 1 and n- 1 / 2 T- 1, depending on whether or not incidental trends are extracted from the panel data. In the latter case, when the alternative hypothesis is homogeneous across individuals, it is shown that the point optimal test and the Ploberger-Phillips test both achieve the power envelope and are uniformly most powerful, in contrast to point optimal unit root tests for time series. Some simulations examining the finite sample performance of the tests are reported.

Original languageEnglish
Pages (from-to)416-459
Number of pages44
JournalJournal of Econometrics
Volume141
Issue number2
DOIs
Publication statusPublished - 2007 Dec 1

Fingerprint

Optimal Test
Unit Root Tests
Envelope
Time series
Panel Data
Testing
Local Power
Asymptotic Power
Unit Root
t-test
Alternatives
Power Function
Trends
Panel unit root tests
Simulation
Optimal test
Power envelope
Panel data

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics

Cite this

Moon, Hyungsik Roger ; Perron, Benoit ; Phillips, Peter C.B. / Incidental trends and the power of panel unit root tests. In: Journal of Econometrics. 2007 ; Vol. 141, No. 2. pp. 416-459.
@article{645e356690a34c139904ad77f98d3e7b,
title = "Incidental trends and the power of panel unit root tests",
abstract = "The asymptotic local power of various panel unit root tests is investigated. The (Gaussian) power envelope is obtained under homogeneous and heterogeneous alternatives. The envelope is compared with the asymptotic power functions for the pooled t-test, the Ploberger and Phillips [2002. Optimal testing for unit roots in panel data. Mimeo] test, and a point optimal test in neighborhoods of unity that are of order n- 1 / 4 T- 1 and n- 1 / 2 T- 1, depending on whether or not incidental trends are extracted from the panel data. In the latter case, when the alternative hypothesis is homogeneous across individuals, it is shown that the point optimal test and the Ploberger-Phillips test both achieve the power envelope and are uniformly most powerful, in contrast to point optimal unit root tests for time series. Some simulations examining the finite sample performance of the tests are reported.",
author = "Moon, {Hyungsik Roger} and Benoit Perron and Phillips, {Peter C.B.}",
year = "2007",
month = "12",
day = "1",
doi = "10.1016/j.jeconom.2006.10.003",
language = "English",
volume = "141",
pages = "416--459",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "2",

}

Incidental trends and the power of panel unit root tests. / Moon, Hyungsik Roger; Perron, Benoit; Phillips, Peter C.B.

In: Journal of Econometrics, Vol. 141, No. 2, 01.12.2007, p. 416-459.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Incidental trends and the power of panel unit root tests

AU - Moon, Hyungsik Roger

AU - Perron, Benoit

AU - Phillips, Peter C.B.

PY - 2007/12/1

Y1 - 2007/12/1

N2 - The asymptotic local power of various panel unit root tests is investigated. The (Gaussian) power envelope is obtained under homogeneous and heterogeneous alternatives. The envelope is compared with the asymptotic power functions for the pooled t-test, the Ploberger and Phillips [2002. Optimal testing for unit roots in panel data. Mimeo] test, and a point optimal test in neighborhoods of unity that are of order n- 1 / 4 T- 1 and n- 1 / 2 T- 1, depending on whether or not incidental trends are extracted from the panel data. In the latter case, when the alternative hypothesis is homogeneous across individuals, it is shown that the point optimal test and the Ploberger-Phillips test both achieve the power envelope and are uniformly most powerful, in contrast to point optimal unit root tests for time series. Some simulations examining the finite sample performance of the tests are reported.

AB - The asymptotic local power of various panel unit root tests is investigated. The (Gaussian) power envelope is obtained under homogeneous and heterogeneous alternatives. The envelope is compared with the asymptotic power functions for the pooled t-test, the Ploberger and Phillips [2002. Optimal testing for unit roots in panel data. Mimeo] test, and a point optimal test in neighborhoods of unity that are of order n- 1 / 4 T- 1 and n- 1 / 2 T- 1, depending on whether or not incidental trends are extracted from the panel data. In the latter case, when the alternative hypothesis is homogeneous across individuals, it is shown that the point optimal test and the Ploberger-Phillips test both achieve the power envelope and are uniformly most powerful, in contrast to point optimal unit root tests for time series. Some simulations examining the finite sample performance of the tests are reported.

UR - http://www.scopus.com/inward/record.url?scp=34248143514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34248143514&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2006.10.003

DO - 10.1016/j.jeconom.2006.10.003

M3 - Article

AN - SCOPUS:34248143514

VL - 141

SP - 416

EP - 459

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -