Using backward means to eliminate individual effects from dynamic panels
- Author
- Gerdie Everaert (UGent)
- Organization
- Abstract
- The within-groups estimator is inconsistent in dynamic panels with fixed T since the sample mean used to eliminate the individual effects from the lagged dependent variable is correlated with the error term. This paper suggests to eliminate individual effects from an AR(1) panel using backward means as an alternative to sample means. Using orthogonal deviations of the lagged dependent variable from its backward mean yields an estimator that is still inconsistent for fixed T but the inconsistency is shown to be negligibly small. A Monte Carlo simulation shows that this alternative estimator has superior small sample properties compared to conventional fixed effects, bias-corrected fixed effects and GMM estimators. Interestingly, it is also consistent for fixed T in the specific cases where (i) T = 2, (ii) the AR parameter is 0 or 1, (iii) the variance of the individual effects is zero.
- Keywords
- Dynamic panel, Individual effects, Backward mean, Monte Carlo simulation, Orthogonal deviations
Downloads
-
wp 553.pdf
- full text
- |
- open access
- |
- |
- 454.28 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-526813
- MLA
- Everaert, Gerdie. “Using Backward Means to Eliminate Individual Effects from Dynamic Panels.” Working Paper Series Faculteit Economie En Bedrijfskunde, vol. 2009/553, Faculteit Economie en Bedrijfskunde, UGent, 2009, pp. 1–22.
- APA
- Everaert, G. (2009). Using backward means to eliminate individual effects from dynamic panels. Ghent: Faculteit Economie en Bedrijfskunde, UGent.
- Chicago author-date
- Everaert, Gerdie. 2009. “Using Backward Means to Eliminate Individual Effects from Dynamic Panels.” Working Paper Series Faculteit Economie En Bedrijfskunde. Ghent: Faculteit Economie en Bedrijfskunde, UGent.
- Chicago author-date (all authors)
- Everaert, Gerdie. 2009. “Using Backward Means to Eliminate Individual Effects from Dynamic Panels.” Working Paper Series Faculteit Economie En Bedrijfskunde. Ghent: Faculteit Economie en Bedrijfskunde, UGent.
- Vancouver
- 1.Everaert G. Using backward means to eliminate individual effects from dynamic panels. Vol. 2009/553, Working Paper Series Faculteit Economie en Bedrijfskunde. Ghent: Faculteit Economie en Bedrijfskunde, UGent; 2009. p. 1–22.
- IEEE
- [1]G. Everaert, “Using backward means to eliminate individual effects from dynamic panels,” Working Paper Series Faculteit Economie en Bedrijfskunde, vol. 2009/553. Faculteit Economie en Bedrijfskunde, UGent, Ghent, pp. 1–22, 2009.
@misc{526813,
abstract = {{The within-groups estimator is inconsistent in dynamic panels with fixed T since the
sample mean used to eliminate the individual effects from the lagged dependent variable is
correlated with the error term. This paper suggests to eliminate individual effects from an
AR(1) panel using backward means as an alternative to sample means. Using orthogonal
deviations of the lagged dependent variable from its backward mean yields an estimator that
is still inconsistent for fixed T but the inconsistency is shown to be negligibly small. A Monte
Carlo simulation shows that this alternative estimator has superior small sample properties
compared to conventional fixed effects, bias-corrected fixed effects and GMM estimators.
Interestingly, it is also consistent for fixed T in the specific cases where (i) T = 2, (ii) the AR
parameter is 0 or 1, (iii) the variance of the individual effects is zero.}},
author = {{Everaert, Gerdie}},
keywords = {{Dynamic panel,Individual effects,Backward mean,Monte Carlo simulation,Orthogonal deviations}},
language = {{eng}},
pages = {{1--22}},
publisher = {{Faculteit Economie en Bedrijfskunde, UGent}},
series = {{Working Paper Series Faculteit Economie en Bedrijfskunde}},
title = {{Using backward means to eliminate individual effects from dynamic panels}},
volume = {{2009/553}},
year = {{2009}},
}