Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Traci Schlesinger <traci.schlesinger@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Goodness of Fit Measures Following xtlogit and xtreg on MI data? |
Date | Wed, 13 Apr 2011 20:31:38 -0500 |
Dear Statalisters: I estimated xtlogit and xtreg models on multiply imputed data (see model below). My data is individual-level data clustered by counties, thus the use of county-level random-effects. A reviewer has asked me to include goodness-of-fit measures in my tables (e.g. a pseudo-R2). I saw the FAQ about calculating goodness-of-fit measures using the log-likelihoods of your constant only and full models. However, i don't get log-likelihoods when using xtlogit command in combination with the mi command. Is there anyway for me to calculate a goodness-of-fit measure after the below model? Thanks! traci mi estimate, dots: xtlogit incarcerate black latino young1 young2 drugt log_tot chg1att second_felony cjstatus priarr prifconv primconv pripris y1998 y2000 y2002 y2004 y2006 if guidelines_cat == 0 * * 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/