[Joao Santos Silva]

Dear Jeff Wooldridge,

Thank you for taking the time to go over what I wrote and to get back to me on this. Indeed, I should have made clear that I was conditioning only on what you called x2m, the exogenous part of x. I also agree that the exogeneity requirements are very different in the two approaches.

Best wishes and, again, thank you for helping me think about this,

Joao

[Susan Jain]

Dear Prof Silva
I am using PPML with time effect and XTPQML with both time and country fixed effects - working on FDI data
Do I need to test endogeneity by using ivpoisson gmm?
I am unable to write this ivpoisson command in stata..
could you please help..

[Joao Santos Silva]

Dear Susan Jain

I am afraid that only you can know whether endogeneity is a potential problem in your model, but for what is worth most papers in the are do not worry about it. As for ivpoisson, please check the help file.

Best wishes,

Joao