Re: st: mfx after svy: probit - "invalid e(rules) matrix" error

["Erin Strumpf" <erinstrumpf@hotmail.com>]

> At 01:28 PM 2/4/2006, Erin Strumpf wrote:
> >   I'd like to calculate marginal effects after running a svy: probit 
> > regression, and since dprobit isn't >>   available for survey commands, 
> > I'm running:
> >
> >    xi: svy, subpop(temp): probit rfullret tx2 i.ragey_e rfemale 
> > i.rraceeth i.raedcat
> >    mfx
> >
> >    the regression runs fine, but then I get the following error:
> >    invalid e(rules) matrix
> >    r(322);
> >
> >    and the mfx command doesn't run. I have searched the Stata help files, 
> > the Stata list, and >>the internet for this mysterious r(322) error, and I 
> > haven't come up with anything. Does anyone >>have any ideas what is going 
> > wrong here?
> >
> >    thanks,
> >    Erin
> . error 322
> something that should be true of your estimation results is not
>
> I don't have your data. But, when I run a different problem with svy: 
> probit I do not encounter >this error. So, I would say
>
> * Make sure you have the latest update of Stata
>
> * Try it without the svy: prefix and see if it works.
>
> * You could also try installing and running the -margeff- command. However, 
> I am not sure if it >works correctly with -svyset- data or not.
I am using Stata 9.1.
If I run the regression without the survey command, the mfx command works 
fine (though obviously my se's are off since I'm not accounting for the 
survey design).
That may have to suffice for now, until I can figure out what an e(rules) 
matrix is....

thanks for the advice,
Erin


*
*   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/