Sargan or Hansen

[ancia dettox]

Is it possible to use hansen test after xtabond? I estimated my model with xtabond....vce(robust) and I want to test validity of my instruments.

Which tests are also better (Sargan or Hansen) after:
1) xtabond
2) xtdpdsys?

Thanks in advance!

[River Huang]

Dear ancia, Please search xtabond2 and install.

[ancia dettox]

Dear ? Why xtabond2? Could you please give me more details?

[River Huang]

Dear ancia, xtabond2 offers both Sargan and Hansen-J statistics.

[Sebastian Kripfganz]

As a bit of shameless self-promotion, xtdpdgmm as well offers both Sargan and Hansen statistics:

• XTDPDGMM: new Stata command for GMM estimation of linear (dynamic) panel data models
• Kripfganz, S. (2019). Generalized method of moments estimation of linear dynamic panel data models. Proceedings of the 2019 London Stata Conference.

The Sargan test is asymptotically valid after one-step difference GMM estimation without vce(robust), assuming that the one-step weighting matrix is indeed optimal, i.e. the idiosyncratic errors are serially uncorrelated and homoskedastic. The Hansen test is asymptotically valid after two-step difference or system GMM estimation without the need to assume homoskedasticity. After one-step difference GMM estimation with vce(robust), the Hansen test is only asymptotitcally valid, if again the one-step weighting matrix is indeed optimal. After one-step system GMM estimation, neither the Sargan nor the Hansen test is asymptotically valid.