[Joao Santos Silva]

Dear Chhavi Tiwari,

Please try the mu option to see if it helps; using fewer variables should also help. Finally, the command is based on Stata's gmm command, so you can use the gmm options to try to get a quicker estimation.

Best wishes,

Joao

[Hamid muili]

Dear Prof Joao Santos Silva , I am trying to use the Ivqreg2 for a model and would be willing to use a GMM based instrument, how do you think this can be modelled.
Poverty= bo +b1POV(t-1) +b2ineq +b3Infla+b4gov_exp+e , instrument (L1.inq,L1.infla,L1.gov_exp) q(0.25 0.5. .75)

[Joao Santos Silva]

Dear Hamid muili,

I am afraid I do not understand your question. What is stopping you form estimating the model?

Best wishes,

Joao

[Joao Santos Silva]

Dear All,

With the usual thanks to Kit Baum, an updated version of ivqreg2 is now available in SSC; the changes are not big, but all users are encouraged to update.

Please do let me know if you find any problems with the new version.

Best wishes,

Joao

[Abdul Abzhelad]

Dear @Joao Santos Silva are there specific commands to be used with ivqreg2 for the autocorrelation, over-identifying restrictions, and joint significance of parameters?

[Joao Santos Silva]

Dear Abdul Abzhelad,

No, there are not.

Best wishes,

Joao

[Abdul Abzhelad]

Thank you for your very prompt answer @Joao Santos Silva.
Bests
Abdul

[Akbar Isayev]

Good day Joao Santos Silva,

I have a question for you related with instruments. As I know, the list of instruments should include all the exogenous variables (so these appear twice in the command) and therefore the number of instruments needs to be at least equal to the number of explanatory variables. What if lets say I have 6 variables and 3 of them is endogenous 3 exogenous. In instruments option if I put only exogenous ones (which is 3), number of instruments wont be equal to the number of explanatory variables. What should be the solution for such a case? Is the only solution is I have to bring some other exogenous instruments from outside of the model?

Thanks in advance

[Joao Santos Silva]

Dear Mugabil Isayev,

Yes, in any instrumental variables estimator you need to have "external" instruments.

Best wishes,

Joao

[Akbar Isayev]

Dear Joao Santos Silva,

Thank you very much for your quick reply and clarification. My second question is: by adding lags of exogenous variables still we can satisfy "number of instruments needs to be at least equal to the number of explanatory variables" condition. Instead of adding external exogenous instruments, can I add lets say first and/or second lags of exogenous variables from my model to satisfy aforementioned condition?

Thank you in advance.

[Joao Santos Silva]

Dear Mugabil Isayev,

If those are valid instruments, then yes.

Best wishes,

Joao

[Akbar Isayev]

Dear Joao Santos Silva,

Thank you very much for your fast response. Last question I have for you. In GMM we have Hansen/Sargan test to check the validity of the instruments. Is there any test like Hansen/Sargan to check the validity of the instruments for ivqreg2? Or should I first apply gmm in order to be sure about the validity of instruments then go for ivreg2 ?

Thanks in advance.

[Joao Santos Silva]

Dear Mugabil Isayev,

The Hansen/Sargan test does not check the validity of the instruments, it is a test of over-identifying restrictions, which is something very different. Anyway, it can be performed in this context, although the command does not provide that result.

Best wishes,

Joao

[Joao Santos Silva]

Dear Mugabil Isayev,

The Hansen/Sargan test does not check the validity of the instruments, it is a test of over-identifying restrictions, which is something very different. Anyway, it can be performed in this context, although the command does not provide that result.

Best wishes,

Joao

[Joana Passinhas]

Dear Joao Santos Silva,

I am using IVQREG2 to estimate structural quantiles but I always get this message:

convergence not achieved
    The Gauss-Newton stopping criterion has been met but missing standard errors indicate some of the parameters are not identified.

Do you know what could be causing it?

Thank you.

Best,
Joana