Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Benjamin Allaire <btallaire@hotmail.com> |
To | "statalist-digest@hsphsun2.harvard.edu" <statalist-digest@hsphsun2.harvard.edu> |
Subject | st: Trouble with margins after suest |
Date | Thu, 30 Jan 2014 12:27:04 -0500 |
Hi Statalist - I'm trying to estimate the marginal effects of a two-part model on inpatient medical expenditures using the delta method. The first part is a logit and the second part is a glm with a log link. I use the suest command to put the two sets of estimates together. The dataset is large (~3.5 mil obs). Every time I try to run margins, it "hangs" in that I get the following message delimiter now ; . margins, dydx(cobst) subpop(if cobst==1) > expression( (exp(predict(equation(inpat_logit_inpat_totpay_dum))) / (1 + exp(predict(equation(inpat_logit_inpat_totpay_dum))))) * > exp(predict(equation(inpat_glm_inpat_totpay))) > ); Warning: cannot perform check for estimable functions. And then.... nothing. When I run this interactively, the indicator bar moves, but it's been running for over a day now and nothing (the glm and logit take about 20 min to run). Any tips on trying to get this to resolve? Thanks, Ben * * 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/