Joakim Westerlund
Professor, Programme director – Master of Data Analytics and Business Economics
The factor analytical approach in near unit root interactive effects panels
Author
Summary, in English
In a recent study, Bai (2013) proposes a new factor analytical (FA) method for estimation of stationary dynamic panel data models with fixed effects. Our interest in this method originates with the fact it does not require explicit demeaning of the data, a practice that is known to cause problems of bias and low power in near unit root panels. The purpose is to study the properties of FA when applied to such panels when the common component admits to a interactive effects representation, which is more general than fixed effects. It is shown that the estimator of the autoregressive parameter is consistent with a well centered asymptotic normal distribution, leading to unit root tests with maximal achievable power. In fact, FA is consistent and asymptotically normal regardless of whether the data are near unit root non-stationary or stationary. It is therefore very general and hence widely applicable.
Department/s
- Department of Economics
Publishing year
2021
Language
English
Pages
569-590
Publication/Series
Journal of Econometrics
Volume
221
Issue
2
Document type
Journal article
Publisher
Elsevier
Topic
- Economics
Keywords
- Bias
- Common factors
- Factor analytical method
- Interactive effects
- Local-to-unity asymptotics
- Panel unit root test
Status
Published
ISBN/ISSN/Other
- ISSN: 0304-4076