RE: st: ivreg2 cannot produce first-stage result

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Phil,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Mo Phil
> Sent: 06 April 2010 03:43
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: ivreg2 cannot produce first-stage result
> 
> Dear Kit,
> 
> Many thanks for your kind advice.
> 
> I changed the command to the following and it worked.
> 
>   ivreg2 my_y $xfirm $year_dummies (choice_dummy = 
> choice_sic3),   first
> 
> By the way, what's the meaning of FSRs?

As Martin says, "first-stage results" or "first-stage regressions".

> The help file of ivreg2 does contain several illustrations of 
> using my previous sepcificaiton of the comand:
> 
> For instance,
>       . ivreg2 lwage exper expersq (=age kidslt6 kidsge6), robust

Your previous specification was:

ivreg2 my_y $xfirm $year_dummies choice_dummy (= choice_sic3),   first

This is not the same as the example in in ivreg2 help file, because of your use of the -first- option.

The difference is that in your previous specification you requested display of the first-stage results, and the example in the ivreg2 help file does not.

This also explains why your previous specification generated an error message.  By saying -first-, you requested display of the first-stage regressions, but there were no first-stage regressions to display because you have no endgenous regressors.

Cheers,
Mark

> Also, one mystery is that I used the same syntax in another 
> project last night, it did work. what's the reason?
> 
> Once again, many thanks.
> 
> Phil
> 
> 
> 
> 2010/4/6 Kit Baum <baum@bc.edu>:
> > <>
> > Phil said
> >
> >
> > I used the following ivreg2 to estimate an IV-regression, but could 
> > not produce the first-stage result.
> >
> > .  ivreg2 my_y $xfirm $year_dummies choice_dummy (= choice_sic3),   
> > first
> >
> > Unable to display first-stage estimates; macro e(first) is missing
> >
> > Unable to display summary of first-stage estimates; macro 
> e(first) is 
> > missing
> >
> >
> > I browsed the Statalist, found a similiar posting, but the 
> suggested 
> > answer is that if OLS is used in the second stage, there is no 
> > first-stage results.
> >
> > But I actually used the same command syntax in another setting some 
> > months ago, it worked.
> >
> > I would appreciated any suggestions on how to obtain the 
> first stage result.
> >
> >
> > The current versions of ivreg2, ivreg29 and ivreg28 from 
> SSC all give 
> > these same two messages. The FSRs that you're considering 
> are the regressions of what lies on the left of the equals 
> sign above on all exogenous variables.
> > As there is nothing there, there are no such regressions to be 
> > estimated. You are using the 'heteroskedastic OLS' (HOLS) 
> estimator, 
> > in which the excluded instruments on the right of the 
> equals sign are 
> > being used to improve the efficiency of the estimates in 
> the presence of heteroskedasticity of unknown form. However, 
> their presence does not change the fact that there are no 
> FSRs in this kind of model.
> >
> > Kit
> > coauthor ivreg2*
> >
> > Kit Baum   |   Boston College Economics & DIW Berlin   |   
> > http://ideas.repec.org/e/pba1.html
> >                              An Introduction to Stata 
> Programming  |   
> > http://www.stata-press.com/books/isp.html
> >   An Introduction to Modern Econometrics Using Stata  |   
> > http://www.stata-press.com/books/imeus.html
> >
> >
> > *
> > *   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/
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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/