Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | vwiggins@stata.com (Vince Wiggins, StataCorp) |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: nbreg & vce(opg) don't work together |
| Date | Tue, 13 Mar 2012 17:09:01 -0500 |
Richard Williams <richardwilliams.ndu@gmail.com> 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 http://www.stata-press.com/data/r12/rod93, 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 vwiggins@stata.com * * 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/