Dear stata users,
I am trying to understand the composition of the instruments used in gmm estimation.
For instance in Arellano and Bond 1991 in table 4 they estimates a gmm for the labor demand. This precise table is reproduce in Roodman 2009 with the following code :
(see page 26)
The result indicate 41 instruments. I guess there is 8 for the regressors and then there is all the available lags for the depend variable. But there is only 9 time periods in the data and 3 are dropped because of the lags in the first stage regression. So only 6 more instruments. My question is then, what are the 27 other instruments ?
Best,
Rémi
References
I am trying to understand the composition of the instruments used in gmm estimation.
For instance in Arellano and Bond 1991 in table 4 they estimates a gmm for the labor demand. This precise table is reproduce in Roodman 2009 with the following code :
Code:
webuse abdata . xtabond2 n L.n L2.n w L.w L(0/2).(k ys) yr*, gmm(L.n) > iv(w L.w L(0/2).(k ys) yr*) nolevel small
The result indicate 41 instruments. I guess there is 8 for the regressors and then there is all the available lags for the depend variable. But there is only 9 time periods in the data and 3 are dropped because of the lags in the first stage regression. So only 6 more instruments. My question is then, what are the 27 other instruments ?
Best,
Rémi
References
- Arellano M, Bond S (1991). “Some Tests of Specification for Panel Data : Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies, 58, 277–297.
- Roodman D (2009). “How to do xtabond2: An introduction to difference and system GMM in Stata.” The Stata Journal, 9, 86-136. https://www.stata-journal.com/articl...article=st0159.
