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]

Re: st: nbreg & vce(opg) don't work together

From (Vince Wiggins, StataCorp)
Subject   Re: st: nbreg & vce(opg) don't work together
Date   Tue, 13 Mar 2012 17:09:01 -0500

Richard Williams <> writes that -nbreg-
does not honor the option -vce(opg)- in Stata 12,

> I am trying to replicate some analyses originally done in Stata 9.
> In Stata 12 I can't get the vce(opg) option to work with nbreg, e.g.
> use, clear
> nbreg deaths i.cohort, nolog
> nbreg deaths i.cohort, nolog vce(opg)
> version 9: xi: nbreg deaths i.cohort, nolog vce(opg)
> The -vce(opg)- is ignored in all cases, i.e. I just get the regular
> standard errors. If I actually fire up my old copy of Stata 9,
> though, the -vce(opg)- option works. Is this a bug that crept into
> Stata somewhere along the way, or am I doing something wrong?

This is indeed a bug.  We will fix it in a future update.

Richard very nicely supplied his own interim workaround, which is to
also include the -technique(bhhh)- option if OPG
(outer-product-of-the-gradient) standard errors are desired.

-- Vince

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index