On the Use of GLS Demeaning in Panel Unit Root Testing
(2018) In Journal of Business & Economic Statistics 36(2). p.309-320- Abstract
- One of the most well-known facts about unit root testing in time series is that the Dickey--Fuller (DF) test based on OLS demeaned data suffers from low power, and that the use of GLS demeaning can lead to substantial power gains. Of course, this development has not gone unnoticed in the panel unit root literature. However, while the potential of using GLS demeaning is widely recognized, oddly enough, there are still no theoretical results available to facilitate a formal analysis of such demeaning in the panel data context. The present paper can be seen as a reaction to this. The purpose is to evaluate the effect of GLS demeaning when used in conjuncture with the pooled OLS $t$-test for a unit root, resulting in a panel analog of the time... (More)
- One of the most well-known facts about unit root testing in time series is that the Dickey--Fuller (DF) test based on OLS demeaned data suffers from low power, and that the use of GLS demeaning can lead to substantial power gains. Of course, this development has not gone unnoticed in the panel unit root literature. However, while the potential of using GLS demeaning is widely recognized, oddly enough, there are still no theoretical results available to facilitate a formal analysis of such demeaning in the panel data context. The present paper can be seen as a reaction to this. The purpose is to evaluate the effect of GLS demeaning when used in conjuncture with the pooled OLS $t$-test for a unit root, resulting in a panel analog of the time series DF--GLS test. A key finding is that the success of GLS depend critically on the order in which the dependent variable is demeaned and first-differenced. If the variable is demeaned prior to taking first-differences, power is maximized by using GLS demeaning, whereas if the differencing is done first, then OLS demeaning is preferred. Furthermore, even if the former demeaning approach is used, such that GLS is preferred, the asymptotic distribution of the resulting test is independent of the tuning parameters that characterize the local alternative under which the demeaning performed. Hence, the demeaning can just as well be performed under the unit root null hypothesis. In this sense, GLS demeaning under the local alternative is redundant. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/8600198
- author
- Westerlund, Joakim LU
- organization
- publishing date
- 2018
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Journal of Business & Economic Statistics
- volume
- 36
- issue
- 2
- pages
- 309 - 320
- publisher
- American Statistical Association
- external identifiers
-
- scopus:85018169857
- ISSN
- 0735-0015
- DOI
- 10.1080/07350015.2016.1152969
- language
- English
- LU publication?
- yes
- id
- b93015d4-58aa-40f5-84a5-7d73cd4ae9ed (old id 8600198)
- date added to LUP
- 2016-04-01 10:12:00
- date last changed
- 2022-03-27 05:57:37
@article{b93015d4-58aa-40f5-84a5-7d73cd4ae9ed,
abstract = {{One of the most well-known facts about unit root testing in time series is that the Dickey--Fuller (DF) test based on OLS demeaned data suffers from low power, and that the use of GLS demeaning can lead to substantial power gains. Of course, this development has not gone unnoticed in the panel unit root literature. However, while the potential of using GLS demeaning is widely recognized, oddly enough, there are still no theoretical results available to facilitate a formal analysis of such demeaning in the panel data context. The present paper can be seen as a reaction to this. The purpose is to evaluate the effect of GLS demeaning when used in conjuncture with the pooled OLS $t$-test for a unit root, resulting in a panel analog of the time series DF--GLS test. A key finding is that the success of GLS depend critically on the order in which the dependent variable is demeaned and first-differenced. If the variable is demeaned prior to taking first-differences, power is maximized by using GLS demeaning, whereas if the differencing is done first, then OLS demeaning is preferred. Furthermore, even if the former demeaning approach is used, such that GLS is preferred, the asymptotic distribution of the resulting test is independent of the tuning parameters that characterize the local alternative under which the demeaning performed. Hence, the demeaning can just as well be performed under the unit root null hypothesis. In this sense, GLS demeaning under the local alternative is redundant.}},
author = {{Westerlund, Joakim}},
issn = {{0735-0015}},
language = {{eng}},
number = {{2}},
pages = {{309--320}},
publisher = {{American Statistical Association}},
series = {{Journal of Business & Economic Statistics}},
title = {{On the Use of GLS Demeaning in Panel Unit Root Testing}},
url = {{http://dx.doi.org/10.1080/07350015.2016.1152969}},
doi = {{10.1080/07350015.2016.1152969}},
volume = {{36}},
year = {{2018}},
}