A Simple Test for Nonstationarity in Mixed Panels: A Further Investigation
(2016) In Journal of Statistical Planning and Inference 173.- Abstract
- The test of Ng (A simple test for nonstationarity in mixed panels, Journal of Business & Economic Statistics 26, 113–126, 2008) is one of the few that enables general inference regarding the proportion of non-stationary units in panel data. The current paper furthers the investigation of Ng (2008) in two directions. First, the existing sequential limit analysis is generalized to a very flexible asymptotic framework in which the number of time periods, T, can be either fixed or tending to infinity jointly with the number of crosssection units, N. Second, the test statistic is evaluated not only under the null hypothesis, but also under alternatives that can be either fixed or local.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/8516076
- author
- Westerlund, Joakim LU
- organization
- publishing date
- 2016-06-01
- type
- Contribution to specialist publication or newspaper
- publication status
- published
- subject
- keywords
- Local asymptotic power, Panel data, Unit root test
- categories
- Popular Science
- in
- Journal of Statistical Planning and Inference
- volume
- 173
- publisher
- North-Holland
- external identifiers
-
- wos:000372557400001
- scopus:84959119502
- ISSN
- 1873-1171
- DOI
- 10.1016/j.jspi.2016.01.004
- language
- English
- LU publication?
- yes
- id
- 453f7f55-79e4-43df-8fcc-fae8502ba689 (old id 8516076)
- date added to LUP
- 2016-04-01 10:09:44
- date last changed
- 2024-06-30 11:00:24
@misc{453f7f55-79e4-43df-8fcc-fae8502ba689, abstract = {{The test of Ng (A simple test for nonstationarity in mixed panels, Journal of Business & Economic Statistics 26, 113–126, 2008) is one of the few that enables general inference regarding the proportion of non-stationary units in panel data. The current paper furthers the investigation of Ng (2008) in two directions. First, the existing sequential limit analysis is generalized to a very flexible asymptotic framework in which the number of time periods, T, can be either fixed or tending to infinity jointly with the number of crosssection units, N. Second, the test statistic is evaluated not only under the null hypothesis, but also under alternatives that can be either fixed or local.}}, author = {{Westerlund, Joakim}}, issn = {{1873-1171}}, keywords = {{Local asymptotic power; Panel data; Unit root test}}, language = {{eng}}, month = {{06}}, publisher = {{North-Holland}}, series = {{Journal of Statistical Planning and Inference}}, title = {{A Simple Test for Nonstationarity in Mixed Panels: A Further Investigation}}, url = {{http://dx.doi.org/10.1016/j.jspi.2016.01.004}}, doi = {{10.1016/j.jspi.2016.01.004}}, volume = {{173}}, year = {{2016}}, }