Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Chris Parker <cparker.phd2007@london.edu> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Multiple imputation combined with collapse |
Date | Mon, 1 Aug 2011 08:08:25 -0400 |
Thanks for the advise Richard. That is definitely counter-intuitive to me as we are losing information to be used in the imputation model but I'll read a bit more to try to understand why this creates biased estimates. Chris On Sat, Jul 30, 2011 at 3:51 PM, Richard Williams <richardwilliams.ndu@gmail.com> wrote: > > At 12:07 PM 7/30/2011, Chris Parker wrote: >> >> Hi Statalist, >> >> I have a question about multiple imputation. I have missing data in one >> of my variables, x in the example below, which I then create several >> passive variables for. I then collapse the data and try to run a >> regression but the collapsed passive variables based on x are identical >> to the collapsed original x variable. Does anyone know what I am doing >> wrong here? I tried searching in [MI] but found no reference to >> -collapse- which I take as a sign that I cannot do what I am trying to >> do. Is -mi- unable to handle collapse? > > I don't know the answer to your question, but I do know that at least some people think passive imputation is a bad idea. Instead, you should compute the variables in your original data set, and then let their missing values be imputed (which is wildly counter-intuitive but seems to work better). See > > http://www.stata.com/statalist/archive/2009-02/msg00602.html > > http://www.stata.com/statalist/archive/2009-02/msg00613.html > > > ------------------------------------------- > Richard Williams, Notre Dame Dept of Sociology > OFFICE: (574)631-6668, (574)631-6463 > HOME: (574)289-5227 > EMAIL: Richard.A.Williams.5@ND.Edu > WWW: http://www.nd.edu/~rwilliam > > * > * 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/ > > ______________________________________________________________________ > > This email has been scanned by the MessageLabs Email Security System > on behalf of the London Business School community. > For more information please visit http://www.messagelabs.com/email > ______________________________________________________________________ * * 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/