Dear members,

In our research, we use the stata command “ivreg2” with year and country dummies in which, after the comma, we type “gmm2s” in order to obtain a two-step efficient GMM estimator. Following Baum et al. (2007), it is not clear to us what type of GMM estimator is applied through the previous command, i.e. difference GMM or system GMM estimator. The former is applied by using the stata command “xtabond” and the latter is applied by using “xtabond2”.

In addition, a minor question also arises: By using “ivreg2” with the same way, the instruments applied in endogenous estimators, are both in levels and difference of variables or not?

Best regards.


References

Baum, C. F., Schaffer, M. E., Stillman, S., 2007. Enhanced routines for instrumental variables / generalized method of moments estimation and testing. STATA Journal 7, 465-506.