I am using the Difference-GMM and System-GMM in my study by applying STATA, where I have reported the test results of the Sargan test, whichi is basically a test for over-identification of instrumental variables.
Hence I have reported the Sargan test results, here my question is that whether again is it required to run the Hansen test.
In case of panel GMM in STATA, Sargan test is provided, so can anybody suggest me in finding the Hansen test for panel GMM in STATA.
I invite some suggestion on these querries.
using xtabond2 (see Roodman 2009) option, gives you a direct results, so you dont have to run the test again ... About using the both of the methods GMM, I recomand for you to read BOND (2002) : 10.1007/s10258-002-0009-9 , it offers a good way to choose between the two methods, good luck
You can compare your results via both the tests may have more impact and clarity to reviewers or readers of your findings/manuscript.
If I am not wrong, the Hansen test is just a generalization when you have more complex data structures.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Energy