GMM is just a class of estimator; an estimator that happens to be naturally well suited to deal with potential endogeneity issues. If u have a cointegrated panel and u want to examine the potential long run relationship between your variables,then as far as I know the most appropriate econometric methodologies are these of Fmols and Dols.
As stated by Dimitrios A. GMM is a class of estimator like OLS. But FMOLS and DOLS is the appropriate method to test for long run relationship in panel cointegration.
Thank you all for the responses.
I undestand that GMM is just a class of estimator which deals with endogeneity issues. FMOLS & DOLS estimator also deals with problems such as endogeneity, heteroskedasticity, serial correlation. So, i was wondering why the GMM (a general estimator) is not appropriate.
Also, If we have lagged endogenous variables in the long run equation?
If you just want to deal with all the aforementioned econometric issues and just to find the link between your y and x's then GMM is fine, but if u want to examine if there is a long run relationship between your variables FMOLS and DOLS are much more appropriate.
Just as Dimitrios noted, it depends on what you intend doing, if you are only interested in long term association fmols and dols should be fine. But when you have a problem of endogeneity to deal with the gmm approach is just appropriate.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing