Bias-Corrected Common Correlated Effects Pooled Estimation in Dynamic Panels
(2021) In Journal of Business & Economic Statistics 39(1). p.294-306- Abstract
- This article extends the common correlated effects pooled (CCEP) estimator to homogenous dynamic panels. In this setting, CCEP suffers from a large bias when the time span (T) of the dataset is fixed. We develop a bias-corrected CCEP estimator that is consistent as the number of cross-sectional units (N) tends to infinity, for T fixed or growing large, provided that the specification is augmented with a sufficient number of cross-sectional averages, and lags thereof. Monte Carlo experiments show that the correction offers strong improvements in terms of bias and variance. We apply our approach to estimate the dynamic impact of temperature shocks on aggregate output growth.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/3f758b0a-b296-4d4a-a0aa-917f341ae56b
- author
- De Vos, Ignace LU and Everaert, Gerdie
- organization
- publishing date
- 2021
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Common correlated effects, Dynamic panel bias, Factor augmented regression, Multifactor error structure
- in
- Journal of Business & Economic Statistics
- volume
- 39
- issue
- 1
- pages
- 13 pages
- publisher
- American Statistical Association
- external identifiers
-
- scopus:85071843335
- ISSN
- 1537-2707
- DOI
- 10.1080/07350015.2019.1654879
- language
- English
- LU publication?
- yes
- id
- 3f758b0a-b296-4d4a-a0aa-917f341ae56b
- date added to LUP
- 2019-09-03 16:13:28
- date last changed
- 2022-04-26 03:52:36
@article{3f758b0a-b296-4d4a-a0aa-917f341ae56b, abstract = {{This article extends the common correlated effects pooled (CCEP) estimator to homogenous dynamic panels. In this setting, CCEP suffers from a large bias when the time span (T) of the dataset is fixed. We develop a bias-corrected CCEP estimator that is consistent as the number of cross-sectional units (N) tends to infinity, for T fixed or growing large, provided that the specification is augmented with a sufficient number of cross-sectional averages, and lags thereof. Monte Carlo experiments show that the correction offers strong improvements in terms of bias and variance. We apply our approach to estimate the dynamic impact of temperature shocks on aggregate output growth.}}, author = {{De Vos, Ignace and Everaert, Gerdie}}, issn = {{1537-2707}}, keywords = {{Common correlated effects; Dynamic panel bias; Factor augmented regression; Multifactor error structure}}, language = {{eng}}, number = {{1}}, pages = {{294--306}}, publisher = {{American Statistical Association}}, series = {{Journal of Business & Economic Statistics}}, title = {{Bias-Corrected Common Correlated Effects Pooled Estimation in Dynamic Panels}}, url = {{https://lup.lub.lu.se/search/files/69080341/CCEPbc_ACCEPTED_AUTHOR_VERSION.pdf}}, doi = {{10.1080/07350015.2019.1654879}}, volume = {{39}}, year = {{2021}}, }