st: RE: RE: RE: Use of aweights command in xtivreg2

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

Dan,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Daniel Brown
> Sent: 07 March 2012 21:19
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: RE: Use of aweights command in xtivreg2
> 
> Mark,
> 
> Thanks. I am still not entirely certain that I have followed.
> 
> I am under the impression from what you have said and from 
> the help file, that areg performs the exact same estimation 
> as xtreg , fe; assuming that I 'absorb' on my panel variable 
> in the areg regression. In other words the only difference 
> between areg and xtreg, fe if you absorb on the panel 
> variable in areg is that areg allows me to use weights that 
> do not have to be constant within the panel over time. But 
> other than that areg , absorb(panelvar) performs fixed 
> effects estimation.

That's right.  You can also check (if you don't have too many dummies)
by using -regress- and including the dummies by hand.

> Yet when I manually time demean variables and run a 
> regression using the weights that vary within the panel (in 
> other words manually doing what I think areg is doing) I get 
> different estimates. So I must have misunderstood somewhere here.

Possibly you are not demeaning correctly.  A common mistake is not to
demean using a consistent sample. BTW, Ben Jann's -center- is a
convenient tool for demeaning by hand.  The -casewise- option will
guarantee you are using a consistent sample.

HTH,
Mark

> Thanks,
> Dan
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu 
> [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Schaffer, 
> Mark E [M.E.Schaffer@hw.ac.uk]
> Sent: 07 March 2012 14:27
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: Use of aweights command in xtivreg2
> 
> Dan,
> 
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Daniel 
> > Brown
> > Sent: 05 March 2012 16:45
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: Use of aweights command in xtivreg2
> >
> > Dear Statalist,
> >
> > I am running weighted regressions using the aweights command (to 
> > reflect the fact that the observations for my dependent 
> variable are 
> > themselves averages calculated on the basis of a varying number of 
> > underlying observations). When I perform fixed effects regressions, 
> > Stata does not allow me to weight observations for the same 
> > cross-sectional unit in different time periods differently. 
> However, 
> > when I then perform IV analysis controlling for fixed effects using 
> > xtivreg2 (,fe), I apparently can weight observations in 
> different time 
> > periods within a cross-sectional unit differently. I wanted to know 
> > why I am able to do this when using xtivreg2 (,fe) but not 
> when using 
> > xtreg (,fe)?
> 
> xtivreg2,fe allows weights because it was easy enough to 
> program the feature - xtivreg2 is just a wrapper for ivreg2.
> 
> But whether or not weighting makes sense is up to the user.
> 
> xtivreg,fe does not allow weights, but areg - which reports 
> the same estimator - does.  The rationale is in -help areg-:
> 
> "areg is designed for datasets with many groups, but not a 
> number of groups that increases with the sample size.  See 
> the xtreg, fe command in [XT] xtreg for an estimator that 
> handles the case in which the number of groups increases with 
> the sample size."
> 
> HTH,
> Mark
> 
> >
> > Thanks,
> >
> > Dan
> > *
> > *   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 the Sunday Times Scottish 
> University of the Year 2011-2012
> 
> We invite research leaders and ambitious early career 
> researchers to join us in leading and driving research in key 
> inter-disciplinary themes. Please see
> 
>     http://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/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 the Sunday Times
Scottish University of the Year 2011-2012

We invite research leaders and ambitious early career
researchers to join us in leading and driving research
in key inter-disciplinary themes. Please see 

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