Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Re: external instruments with xtabond2


From   San K <devank@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Re: external instruments with xtabond2
Date   Sun, 15 Apr 2012 21:38:41 +1000

I'm also learning xtabond2  too. I also had similar question but chose ignore.
Anyway, my understanding (not that great) is that the estimator is
going use the instruments to predict the endogenous variable and then
the predicted variable is used in the model.
May be (I have no basis), the estimator uses all  iv() and gmm( ) to
predict a value and then that value is used?




On Sun, Apr 15, 2012 at 8:51 PM, Joe Bond <joe.a.bond@googlemail.com> wrote:
> Dear Statalist,
>
> I am trying to run some regressions using the xtabond2 command for
> difference GMM estimation. I have an endogenous explanatory variable
> in my model for which I have found an external instrument; I also have
> other endogenous explanatory variables (including the lagged dependent
> variable) for which I will need to use internal instruments; and
> finally I have a couple of strictly exogenous explanatory variables..
>
> At the moment I am including all endogenous regressors in the gmm( )
> section and all exogenous regressors and the extermal instrument in
> the iv() section. I was wondering, will this arrangement work as
> intended? What I don't quite get is how Stata will know to use the
> external instrument for the specific endogenous variable I intend it
> to be used for, and also whether I should be putting this variable in
> the gmm() section when I want it to be instrumented using the external
> variable rather than its own lags. Would I be better off just using
> the external instrument on this variable separately beforehand and
> then running the xtabond2 command afterwards using the predicted
> values from this regression?
>
> Any help would be much appreciated.
>
> Joe
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index