Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Antwort: st: interactive Wald test?

From   Richard Ochmann <>
Subject   Antwort: st: interactive Wald test?
Date   Thu, 22 Jul 2010 18:59:08 +0200


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 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:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index