Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Srinivas <geevasan@yahoo.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: 95%CrI |
Date | Sun, 11 Nov 2012 19:17:18 -0800 (PST) |
Dear Dirk Many thanks that was helpful. rgds srinivas --- On Thu, 11/8/12, Dirk Enzmann <dirk.enzmann@uni-hamburg.de> wrote: > From: Dirk Enzmann <dirk.enzmann@uni-hamburg.de> > Subject: Re: st: 95%CrI > To: statalist@hsphsun2.harvard.edu > Date: Thursday, November 8, 2012, 3:31 AM > Presumably, Srinivas refers to the > credible (confidence) intervall of the MOR as discussed in > Merlo et al. (2006), see > > http://www.stata.com/statalist/archive/2011-05/msg00670.html > > As far as I know, this is not (yet) possible using Stata. > When asking Juan Merlo directly, his answer was: > > "The 95% credible interval is based on the Bayesian MCMC > estimation. You use the extreme 2.5% of the posterior > distribution of the second level variance and calculate the > MOR for these limits. This is easy to obtain in MLwiN and I > think now you can run the MLwiN within STATA. Hope my > explanation helps you." > > See also -findit runmlwin- . > > Surely, there is an absurdity to run a multilevel model > using Stata, run an .ado-file (see -findit xtmrho-) to > calculate the MOR, and then run the same model again in > MLwiN to obtain its 95% credibility interval. > > Dirk > > At 08.11.2012 08:33, Srinivas<geevasan@yahoo.com> > wrote: > > Hi > > can anyone tell me the stata command or module to > calculte 95% Credible Interval (95%CrI) for Median odds > Ratio > > and also > > stata command or module to calculte IOR-80 and Sorting > out index. > > > > your time an dhelp si appreciated > > ======================================== > Dr. Dirk Enzmann > Institute of Criminal Sciences > Dept. of Criminology > Rothenbaumchaussee 33 > D-20148 Hamburg > Germany > > phone: +49-(0)40-42838.7498 (office) > +49-(0)40-42838.4591 (Mrs > Billon) > fax: +49-(0)40-42838.2344 > email: dirk.enzmann@uni-hamburg.de > http://www2.jura.uni-hamburg.de/instkrim/kriminologie/Mitarbeiter/Enzmann/Enzmann.html > ======================================== > * > * 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/ > * * 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/