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 at the end of May, and its replacement, statalist.org is already up and running.


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

st: RE: xtoverid with the partial() option?


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: xtoverid with the partial() option?
Date   Tue, 22 Nov 2011 23:34:32 -0000

Jacob,

Modifying xtoverid to have a partial option would be a hassle, esp. if
you're not used to programming ados (let alone someone else's ados).

It's not very appealing, but you could do G2SLS by hand and then call
ivreg2 yourself to do the RE estimation.  xtivreg reports the estimated
variances of the error components, and you can do the quasi-demeaning
etc. yourself.

Is there a partialling-out strategy that you could use to transform your
variables to remove the uninteresting dummies before calling xtivreg or
xtoverid?

--Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Jacob Fowles
> Sent: 21 November 2011 22:42
> To: statalist@hsphsun2.harvard.edu
> Subject: st: xtoverid with the partial() option?
> 
> All,
> 
> I am using the trick that has been discussed here on numerous 
> occasions of combining xtivreg, re followed by the 
> user-written xtoverid with the noisily and cluster() options 
> to estimate a random-effects IV regression with 
> cluster-robust standard errors.  My model includes a set of 
> dummy variables that are important to include as controls but 
> I do not particularly care about their individual 
> coefficients, and their presence seems to be preventing the 
> covariance matrix of moment conditions from being full rank 
> so overidentification tests are not reported.  Is it somehow 
> possible to modify xtoverid's internal call to ivreg2 to 
> include the partial() option to suppress the estimation of 
> coefficients for these dummy variables?  I have toyed with 
> xtoverid's ado-file to attempt to do this, but the fact that 
> xtoverid relies upon the Stata-assigned internal variable 
> names rather than the names that appear in my dataset have 
> left me puzzled as to how this can be accomplished.
> 
> Potentially relevant:  I'm running Stata 12/SE in Windows 7 
> (64 bit) and have installed all recent Stata updates.  The 
> latest versions of all relevant user-written ado-files have 
> been installed.
> 
> Best,
> 
> Jacob
> *
> *   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.

Heriot-Watt University is the Sunday Times
Scottish University of the Year 2011-2012



*
*   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