This paper extends the half-panel jackknife (HPJ) estimator to GMM models with fixed effects. The Monte Carlo results show that the HPJ significantly reduces finite-sample bias for both the difference… Click to show full abstract
This paper extends the half-panel jackknife (HPJ) estimator to GMM models with fixed effects. The Monte Carlo results show that the HPJ significantly reduces finite-sample bias for both the difference and system GMM estimators of the dynamic panel model.
               
Click one of the above tabs to view related content.