I used IVREG2H to estimate the following
Y= endog(X) exog(Z)
I used both 2sls and gmm2s estimation methods and on the endogenous variable of interest I found a slight but persistence difference.
I tested with Hausman to check the significance of the difference but it not significant.
Lewbel (2012) affirmed that GMM estimation is more efficient than 2sls. But Baum et al. (2012)* said that gmm2s is not equivalent to the Lewbel (2012) gmm.
How I can decide which is the best estimator?
*http://fmwww.bc.edu/repec/bocode/u/UK12_baum.pdf (p. 44/51)
Thanks for any help
Y= endog(X) exog(Z)
I used both 2sls and gmm2s estimation methods and on the endogenous variable of interest I found a slight but persistence difference.
I tested with Hausman to check the significance of the difference but it not significant.
Lewbel (2012) affirmed that GMM estimation is more efficient than 2sls. But Baum et al. (2012)* said that gmm2s is not equivalent to the Lewbel (2012) gmm.
How I can decide which is the best estimator?
*http://fmwww.bc.edu/repec/bocode/u/UK12_baum.pdf (p. 44/51)
Thanks for any help
