[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: RE: ivprob and collinearity
There is a -collinear- option for -ml- that would stop collinear
variables from being dropped, but I can't see how to get -probit- to
The alternative is either to remove collinear variables by hand
before you estimate, or add a line early on in -ivprob- that removed
the collinear variables from `exog'. Something like
_rmcoll `exog' if `touse'
local `exog' `r(varlist)'
Hope this helps.
Date sent: Thu, 9 Sep 2004 19:38:48 +0930
From: Patricia Sourdin <email@example.com>
Subject: Re: st: RE: ivprob and collinearity
Send reply to: firstname.lastname@example.org
> Thanks Nick - you are quite right. However, there seemed to be no outcome. I
> have emailed Joe Harkness and will wait for his reply. If you can help in the
> Quoting Nick Cox <email@example.com>:
> > I think this has been discussed previously on Statalist.
> > I can't recall the outcome but you could try searching
> > the archives.
> > Nick
> > firstname.lastname@example.org
> > Patricia Sourdin
> > > I have a query on the ivprob command.
> > > After having set trace on I discovered where I got the
> > > conformability error
> > > message - it is after the last line of this bit of program
> > >
> > >
> > > /* get lee model estimates */
> > > cap estimates drop lastpbt
> > > probit `depvar' `fitted' `exog' if `touse'
> > > matrix tmp = get(_b)
> > > matrix beta=tmp[1,1..`nendog']
> > > matrix means=get(mns)
> > > matrix colnames means = `endog' `exog' _cons
> > >
> > > The probit part of the command automatically excludes
> > > collinear variables and
> > > therefore the 'exog' list and the list of regressors are
> > > different so that
> > > when stata tries to assign column names - there are less
> > > columns than names in
> > > the varlist.
> > >
> > > Presumably this problem will reoccur at later stages of the program.
> > > Can anyone suggest a simple way of amending this program to
> > > deal with this
> > > problem? My programming skills are not so great.
> > *
> > * For searches and help try:
> > * http://www.stata.com/support/faqs/res/findit.html
> > * http://www.stata.com/support/statalist/faq
> > * http://www.ats.ucla.edu/stat/stata/
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
Prof. Mark E. Schaffer
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3485 CERT administrator
* For searches and help try: