Dear Statalist Community,

I want to estimate a dynamic panel data model with fixed effects. The model takes the following form:

Y_it = theta*Y_i,t-1 + beta*Treatmentdummy_it + psi*X_it+alpha_i + delta_t + eps_it

I have an unbalanced panel dataset with predetermined regressors, and my sample size is: N=13, T= 25 on average.

I want to use the bias corrected estimator developed by Hahn and Kuersteiner (2002). However, they discuss 2 cases. The second case is very easy to apply as the bias corrected estimator is given by thetahat2 = (T+1)/T * thetahat1 + 1/T. However, this case is applicable when the model is a univariate stationary panel AR(1) model with fixed effects. As such, I believe it is not appropriate for my case and would therefore have to use the first case discussed by the authors.

The bias corrected estimator in the more general case is given by formula (3) in their paper. However, the formula is complex and I do not understand why it is in vector form and how to apply it.

Therefore, I would appreciate any help regarding how to apply the formula in Stata. Also, please let me know if my understanding is wrong and the first case is actually appropriate for my model.

The paper can be found here: https://www.jstor.org/stable/pdf/308...9GS7AJcs6DFiqQ