Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Mashhood Sheikh <senor_massao@hotmail.com> |
| To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
| Subject | st: Advice needed about convergence problem in –binreg- |
| Date | Mon, 6 Jan 2014 15:23:02 +0000 |
Hi, I need some advice about the convergence problem in using log-binomial regression in stata. The aim is to estimate RR when the outcome is not rare. I am using this command: binreg y t x1 x2 x3 x4 x5 i.x6, rr The model does not converge. One solution I am using is predicting the propensity scores, making a quintile variable out of the scores, and using that in the -binreg- model instead. Is it correct to do so? This is what I am doing: logistic t x1 x2 x3 x4 x5 i.x6 predict prop_score xtile ps_quintiles = prop_score, nq(5) binreg y t i.ps_quintiles, rr Any advice is welcomed. Thankfully, Mashhood Ahmed University of Tromsø, Norway * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/