The LSDV estimator is known to be strongly biased in dynamic panels. The goal of this project is to investigate the performance of a boostrap-based bias correction for models with higher order dynamics and a vector of (endogenous) explanatory variables. Furthermore, we explore alternative ways to estimate the long-term impact. Finally, we want to apply our findings to the convergence debate.
Linear regression is the most famous type of regression analysis in statistics. A statistical analysis of a linear regression model usually begins with estimation of the regression coefficients and continues with measuring the accuracy of the estimators. Unfortunately, it is well known that a traditional statistical analysis based on the least squares principle is very sensitive to outliers in the data. Although many robust estimators have ...
Bias correcting panel data estimators in dynamic models with cross-sectional dependence and endogeneity Ghent University
The FE and CCEP estimators show considerable bias in dynamic models. Many corrections have been devised for FE, but none are applicable to case of endogenous regressors. We address this issue with the bootstrap method. The CCEP estimator is essential under cross-section dependence but has not yet been bias corrected. Such corrections (analytical and bootstrap-based) constitute our second research goal.