Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: ivprob and collinearity


From   "Mark Schaffer" <M.E.Schaffer@hw.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: ivprob and collinearity
Date   Thu, 09 Sep 2004 11:40:36 +0100

Patricia,

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 
take it.

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)'

might work.

Hope this helps.

--Mark

Date sent:      	Thu,  9 Sep 2004 19:38:48 +0930
From:           	Patricia Sourdin <patricia.sourdin@adelaide.edu.au>
To:             	statalist@hsphsun2.harvard.edu
Subject:        	Re: st: RE: ivprob and collinearity
Send reply to:  	statalist@hsphsun2.harvard.edu

> 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 
> meantime...
> thanks
> Patricia
> 
> Quoting Nick Cox <n.j.cox@durham.ac.uk>:
> 
> > I think this has been discussed previously on Statalist. 
> > I can't recall the outcome but you could try searching
> > the archives. 
> > 
> > Nick 
> > n.j.cox@durham.ac.uk 
> > 
> > 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
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS  UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index