xtabond2

[Bruno Moreira]

Hello everyone,

I am currently working with dynamic panel data. The model I want to estimate using the Arellano-Bond estimator is the following:
logYi,t = αi + β0Si,t + γlogYi,t−1 + β1Si,t−1 + εi,t

where Yi,t is the logarithm of GDP per worker in country i (lrgdpw) in year t and S is educational attainment in years (edu25). We treat education as strictly exogenous. The variable t in the dataset identifies the year.


How exactly do I use the xtabond2 command to estimate the model? As in this case, I already have the lags of the variables in my model, I am unsure how I should use the command.

Thank you in advance!

BR,

Bruno

[Bruno Moreira]

I've tried the following command but I feel like it is wrong:

xtabond2 lrgdpw l.lrgdpw edu25 l.edu25, gmm(l.lrgdpw, eq(d)) iv(edu25 l.edu25, eq(d)) robust twostep

Any help is much appreciated!

[Sebastian Kripfganz]

Your command line appears to be correct.

More on the 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.