xtabond2: Observing Y for a longer period than X

[Hulya Kocyigit]

Hello,

I am using the xtabond2 command to run a GMM-IV regression on the equation

Y_{it}=a+b_1Y_{it-1}+b_2 X_{it}+u_{i}+\epsilon_{it}

X is the main treatment variable of interest. Both Y_{it-1} and X_{it} are endogenous so I construct GMM-type instruments for them.

I observe X only for 2015-2020 but I observe Y for 2010-2020. While constructing GMM-type instruments for Y, I use all the available information from 2010-2020. Is there anything wrong with this, should I keep Y only for 2015-2020?

Thank you!

[Sebastian Kripfganz]

There is nothing wrong with this.

More on 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.