I am using a panel dataset of 26 countries for the period 1995-2015. I have a lagged dependent variable in my regression equation. I am confused whether I should use a difference-GMM (with the command -xtabond-) or the system-GMM (command -xtabond2). Can somebody please guide as to which one should I choose given my sample size?
-
Login or Register
- Log in with
