Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Multiple Imputation on Panel Data: all variables have missing data, and the panels are expanding


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Multiple Imputation on Panel Data: all variables have missing data, and the panels are expanding
Date   Tue, 24 Aug 2010 10:04:26 +0000 (GMT)

--- On Tue, 24/8/10, Jennie Day wrote:
> 1)  I am trying to use STATA 12's mi commands to do
> multiple imputations on a panel dataset where all variables
> have some missing values (save the unique ID numbers and the
> time variable).

This worries me alot. Remember that what multiple imputation 
tries to do is to replicate the observed patterns of association
onto the missing values. If none of the observations are complete
then there is no observation that provides information on the
complete set of associations. I would look long and hard at the
pattern of missing values and look at which associations can be 
estimated in this data and which cannot. Based on that I would
try to figure out what kind of imputation model would be possible,
and if you would consider that model acceptable. It may very well
be that there is no acceptable solution for your data. It may
also very well be that you'll need to program your own special
purpose imputation model.

Hope this helps,
Maarten

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

http://www.maartenbuis.nl
--------------------------


      

*
*   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   |   Site index