Re: st: Obtaining scores after clogit

[Karyen Chu <k_chu@uclink.berkeley.edu>]

Thank you very much to both Bobby Gutierrez and Joseph Coveney for your 
helpful suggestions.

I have run Bobby Gutierrez's clogit_score on a small sample of my data and 
it does exactly what I need.

Thank you again

Karyen


At 11:38 AM 7/28/2004 -0500, you wrote:

> Karyen Chu <k_chu@uclink.berkeley.edu> asked:
>
> > I have run a McFadden's choice model using clogit and cannot find any 
> method
> > for obtaining the score matrix since clogit is one of the commands without
> > the score( ) option.
>
> > I have consulted section 23.15 "Obtaining Scores" in the user manual and
> > looked at and tried the commands estimates and ereturn.  So far, 
> nothing has
> > worked.
>
> > I need the scores to perform a score test because the McFadden's Choice
> > model that I have estimated is actually the constrained form of a more
> > complicated model and I am trying to test the constraints using a score 
> test
> > since the unconstrained model will be a bear to estimate.
>
> Joseph Coveney <jcoveney@bigplanet.com> replied:
>
> > Jeroen Weesie wrote a conditional logistic regression command, -clogit2-,
> > that provides scores in order to work with his -suest-, which requires
> > scores from the estimation commands.
>
> From looking at -clogit2-, it appears to work only with one-success-per-group
> estimation, which may be fine since Karen is fitting a choice model.
>
> To obtain scores in the more general case, take a look at my -clogit_score-
> command.
>
>    . net from http://www.stata.com/users/rgutierrez
>
> --Bobby
> rgutierrez@stata.com
> *
> *   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/
Karyen Chu
Dept of Economics
U.C. Berkeley

*
*   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/