Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: st: Working With mi impute mvn for missing data |
Date | Tue, 26 Apr 2011 00:34:00 +0100 |
The t-test is a special case of regression. No need to re-invent the wheel. This was the subject of a recent thread. Nick On Mon, Apr 25, 2011 at 11:24 PM, Clifton Chow <clifton_chow@post.harvard.edu> wrote: > Dear List, > > I have a dataset with 7 ordinal variables of interest that are MAR (Missing at Random) but NOT monotone. I am proceeding with imputation using mi impute mvn VARLIST method with Add (9). The result I get is 9 imputations of my 7 variables of interest from observation that are missing on those variables. I need to determine if the mean average on those variables differ significantly for two cohorts, but while there is mi estimates regress there is no apparent mi estimates ttest. Should I average the means of all 9 imputations and conduct the traditional ttest? > * * 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/