RE: st: RE: margeff after the glm command

["Nick Cox" <n.j.cox@durham.ac.uk>]

Is this a case of "Go Firth and multiply?" 

Nick 
n.j.cox@durham.ac.uk 

Martin Weiss

To get good answers, I think you should show the exact sequence of commands
leading up to the error. I do get results with -glm- and -ssc d margeff-:

***
webuse lbw, clear
glm low age lwt race smoke ptl ht ui, family(binomial)
margeff
***

Also,

***
webuse hiv1, clear
cap which firthlogit
if _rc ssc inst firthlogit
firthlogit hiv cd4 cd8
margeff
***

leads to a much clearer error message than the one you reported...

HTH
Martin


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Sabrina Carrossa
Sent: Donnerstag, 13. Mai 2010 17:53
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: RE: margeff after the glm command

Yes Martin, the e(depvar) command works (sorry), but I still have
problem in executing the margeff command.

On 13 May 2010 16:48, Martin Weiss <martin.weiss1@gmx.de> wrote:
>
> <>
>
> You would have to type
>
> *************
> di e(depvar)
> *************
>
> in any case to make this work...
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Sabrina
Carrossa
> Sent: Donnerstag, 13. Mai 2010 17:16
> To: statalist@hsphsun2.harvard.edu
> Subject: st: margeff after the glm command
>
> Hi all,
>
> I wish to compute the average marginal effect (margeff) after the glm
> command.
>
> Indeed, I have a separation problem, so I found  the "firthlogit
> program" writtend by Joseph Coveney (thanks!):
> http://www.stata.com/statalist/archive/2005-11/msg00786.html
>
> Unfortunately stata shows the following error when I try to execute
> the margeff command: "__000006 not found
> r(111)"
>
> I guess that the problem is that stata doesn't find the dependent
> variable, becouse when I type: "e(depvar)" it shows: "depvar not found
> r(111)";
>
> Is there any chance to use the logit postest command after the glm
> (with the link option?)
>
> I hope it is clear.
> Thanks for your help!
>

*
*   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/