Re: Antwort: st: interactive Wald test?

[Jeph Herrin <stata@spandrel.net>]

Thanks - that was it, sort of. My matrix V had the correct
row names, but the column names were generic.

cheers,
Jeph

On 7/22/2010 12:59 PM, Richard Ochmann wrote:
> ...
>
> Every time when I had the error message "name conflict" after -ereturn
> post-, the reason always was a conflict in the colnames and rownames of
> the matrices replacing and to be replaced.
> Watch out that these exactly coincide over these matrices.
>
> best, rich
>
>
> owner-statalist@hsphsun2.harvard.edu schrieb am 22.07.2010 17:32:47:
>
> > Someone has given me some parameter estimates - including the variance/
> > covariance matrices  - for a number of models. I would like to
> > perform a few Wald tests on groups of independent variables, in the
> > manner of -testparm-.  I know the formula for a Wald test and how
> > to go about programming this, but would surely prefer to find a
> > pre-existing solution. First question, does someone know one?
> >
> > My proposed solution, which may not be the best, is to store my
> > matrices b and V in e(b) and e(V), via -ereturn post-, and go
> > from there. However, this always returns an error:
> >
> >    . ereturn post b V
> >    name conflict
> >    r(507);
> >
> > -ereturn list- shows no stored estimates; using simply -ereturn post-
> > with no arguments gives no error but also creates no estimates.
> > Thoughts on what the name conflict is, and what I am doing wrong?
> >
> > thanks,
> > Jeph
> >
> >
> > *
> > *   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/
>
> *
> *   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/
*
*   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/