Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Clyde B Schechter <clyde.schechter@einstein.yu.edu> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: Re: st: Situation where multiple imputation may be of no use? |
Date | Fri, 10 Feb 2012 20:08:50 +0000 |
Thanks to Richard Williams for his thoughts. The vonHippel 2007 reference linked in his reply is particularly helpful and completely explains the theory underlying my intuitions: the observations with missing dependent variable contain no information about the regression coefficients, and, in fact, are in most situations best omitted from the MI analysis. Well, I was hoping I was wrong, because it might have been an easy way to save our project. Instead, we'll have to do some hard work figuring out something else. Clyde Schechter Department of Family & Social medicine Albert Einstein College of Medicine Bronx, NY, USA * * 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/