Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Missing indicator method for partially mising baseline measurements in randomized trials


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Missing indicator method for partially mising baseline measurements in randomized trials
Date   Fri, 17 Jul 2009 16:43:45 +0000 (GMT)

--- On Fri, 17/7/09, Etan Lakam wrote:
> I am analysis trial data from a randomized trial, 
> comparing the the body mass index (bmi2)  bewteen 
> the grousp at follow-up adjusting for the baseline
> values of bmi (bmi1) Some of the values of bmi at
> baseline are missing and I was advised to use the
> missing indicator method to account for the
> missing values of bmi at baseline, thus to have a
> more efficient result. Being a movice in the field
> of missing data, I don not have a clue on how to
> do this is Stata.

That is a good thing, because you should not do it.
this is explained here:
http://www.stata.com/statalist/archive/2007-12/msg00030.html

If you really care about those observations and
your missing data process satisfies the MAR 
assumption you could use -ice-. To install 
-ice- type in Stata -findit ice- and follow
the links. (The MAR assumption means that the 
probability of being missing does not depend 
on the unobserved values, this assumption can 
thus only be made plausible through a 
theoretical argument and can never be 
emprically tested)

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index