Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

RE: st: RE: xtivreg2 with lagged variables


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: xtivreg2 with lagged variables
Date   Tue, 22 Oct 2013 18:15:22 +0000

Giacomo,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Giacomo Pallaro
> Sent: 22 October 2013 18:12
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: xtivreg2 with lagged variables
> 
> Mark,
> I tried to perform the two stages separately, using twice - xtreg,fe .
> 
> xtreg L.endog_var L.instrument1 L.instrument2 L.married L.age L.age2, fe
> predict instrumented_var
> xtreg depvar L.instrumented_var married age age2, fe
> 
> Doing so I have the advantage to do what I want but I lose the first
> stage post-estimation statistics that are automatically provided by
> xtivreg2.

Another drawback is that the SEs reported in the second stage will be wrong.

> That's why I asked for a way to reach the result as in the
> case of  two saparated stages but using xtivreg2.
> If I have well understood this is not feasible, isn't it?

Not exactly.  As the Stata FAQ I cited makes clear, it's not that it's technically not feasible, it's that it's usually a bad idea.  But if you are really sure you want to go down this route anyway, you might be able to get -sem- to it, since basically you are talking about estimation with 2 structural equations.

--Mark

> 
> Giacomo
> 
> On Tue, Oct 22, 2013 at 6:48 PM, Schaffer, Mark E <M.E.Schaffer@hw.ac.uk>
> wrote:
> > Giacomo,
> >
> > This is actually a general IV question, and the answer is the subject of a
> StataCorp FAQ:
> >
> > http://www.stata.com/support/faqs/statistics/instrumental-variables-
> regression/
> >
> > Short answer: you probably don't want to do it!
> >
> > --MS
> >
> >> -----Original Message-----
> >> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> >> statalist@hsphsun2.harvard.edu] On Behalf Of Giacomo Pallaro
> >> Sent: 22 October 2013 15:43
> >> To: statalist@hsphsun2.harvard.edu
> >> Subject: st: xtivreg2 with lagged variables
> >>
> >> Dear all,
> >>
> >> I'm running on Stata 12 some IV regressions using xtivreg2 on panel data
> and I
> >> am wondering whether it is possible to avoid the varlist1 variables to be
> >> employed into the first stage.
> >> If I use
> >>
> >> xtivreg2 depvar married age age2 (L.endog_var = L.instrument1
> >> L.instrument2), fe
> >>
> >> I get
> >> First stage: L.endog_var on L.instrument1 L.instrument2 married age age2
> >> Second stage: depvar on L.instrumented_var married age age2
> >>
> >> but for me it is pointless to have - married age age2 - in the first stage while
> >> predicting a variable that is one period back in time. I wuold like to have -
> >> L.married L.age L.age2 - to be coherent with L.endog_var.
> >> The second stage instead is correct as it is since the endogenous variable
> shall
> >> be lagged in time with respect to the depvar.
> >>
> >> I tried to run
> >>
> >> xtivreg2 depvar married age age2 (L.endog_var = L.instrument1
> >> L.instrument2  L.married L.age L.age2), fe
> >>
> >> what comes out is that the second stage is the same as before, while the
> first
> >> stage becomes:
> >> L.endog_var on L.instrument1 L.instrument2 married age age2 L.married
> L.age
> >> L.age2
> >>
> >> and unfortunately is not what I need.
> >> What can I do to reach my goal?
> >>
> >> Best regards,
> >> Giacomo Pallaro
> >> *
> >> *   For searches and help try:
> >> *   http://www.stata.com/help.cgi?search
> >> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> >> *   http://www.ats.ucla.edu/stat/stata/
> >
> >
> > -----
> > Sunday Times Scottish University of the Year 2011-2013
> > Top in the UK for student experience
> > Fourth university in the UK and top in Scotland (National Student Survey 2012)
> >
> > We invite research leaders and ambitious early career researchers to
> > join us in leading and driving research in key inter-disciplinary themes.
> > Please see www.hw.ac.uk/researchleaders for further information and how
> > to apply.
> >
> > 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/faqs/resources/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/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/


----- 
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)


We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index