GMM estimation using the xtdpdgmm command

[Abdulwahab Bello]

Dear scholars,
I am trying to estimate a system-GMM using the xtdpdgmm command. My dependent variable is co2 while my independent variables are; ict_ind re eu gdp gdp2 (gdp square) edu trad urb. I specify the model as; xtdpdgmm L(0/1). co2 ict_ind re eu gdp gdp2 edu trad urb, collapse gmm(co2, lag(2 4) model(diff)) gmm( ict_ind re eu gdp gdp2 edu trad urb, lag(1 3) model(diff)) ///
gmm(co2, lag(1 1) diff) gmm(ict_ind re eu gdp gdp2 edu trad urb, lag(0 0) diff) two vce(r)

Please is the above specification correct?

[Sebastian Kripfganz]

With your code, you are estimating the model by system GMM, treating all independent variables as predetermined. If that is your intention, then the code appears to be suitable. You can subsequently check with the specification with the Arellano-Bond test for serial correlation (estat serial) and the Hansen overidentification test (estat overid). You may want to include time dummies, which is common practice when you have macroeconomic data.

• Kripfganz, S. (2019). Generalized method of moments estimation of linear dynamic panel data models. Proceedings of the 2019 London Stata Conference.