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]

st: RE: problems combining xtivreg and xtoverid or using xtivreg2 with random effects estimation


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: problems combining xtivreg and xtoverid or using xtivreg2 with random effects estimation
Date   Fri, 30 Sep 2011 15:52:24 +0100

Daniel,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Daniel Schalling
> Sent: Friday, September 30, 2011 11:10 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: problems combining xtivreg and xtoverid or using 
> xtivreg2 with random effects estimation
> 
> > But this is peculiar.  If you estimate using -xtivreg- but 
> exclude all the
> variables that Stata is dropping, -xtoverid- should now work.  
> 
> I tried it again and again and somehow it worked now. I 
> excluded all the
> omitted variables and xtoverid worked. Thank you very much for that!
> 
> > was introduced in Stata 11, if I recall.  It's possible that if you
> precede your -xtivreg- estimation with the line version 10 [...]
> 
> I tried this as well, because I'm using Stata 11. It seems to 
> solve the
> problem with the omitted variables, too. But xtoverid brings 
> out another
> error message: " xtoverid error: internal reestimation of eqn 
> differs from
> original r(198);".

My guess is that you have some perfectly collinear regressors, and some
have to be dropped.  xtivreg is dropping one set, and xtoverid is
dropping a different set, and hence the internal restimation by xtoverid
doesn't match the original done by xtivreg.

> But this doesn't matter, due to the fact 
> that xtoverid
> now works.
> 
> I have another question concerning this topic: Do you have any idea or
> advice for me, how to get similar postestimations for xtivreg 
> like the ones
> you get in xtivreg2: weak identification test and 
> underidentification test?

xtoverid has an undocumented option, noi (or noisily).  If you say

xtoverid, noi

it will report the internal reestimation (done by ivreg2) including the
weak and under identification stats.

Cheers,
Mark


> 
> A few information about my analysis:
> - a panel data with 62 groups and each 60 observations
> - 59 exogenous variables, a few of them are dummies
> - 2 endogenous variables with  5 instruments
> 
> Thank you all a lot, Daniel.
> 
> 
> *
> *   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