--- joe ragg <[email protected]> wrote:
> I have a multiple logistic regression model, where the predictor
> variables are binary. I wanted to impute values for missing predictor
> variables so as to make full use of the available data. The stata
> impute command, as you all may be aware, imputes a value that is
> calculated according to a multiple regression model of the other
> predictor variables on the predictor variable of interest, and so the
> imputed values are not binary but rather between 0 and 1. I was
> wondering if this is a valid technique given my binary predictor
> variables. I would like some assurance that it is valid, as I think
> that probably it is valid.
Unfortunately, -impute- is not valid at all. A long discussion of this
point was started here:
http://www.stata.com/statalist/archive/2008-02/msg00550.html
In short you should use -ice-, see: -findit ice-.
Royston, P. 2004. Multiple imputation of missing values. Stata Journal
4(3): 227�241.
Royston, P. 2005. Multiple imputation of missing values: update. Stata
Journal 5(2): 188�201.
Royston, P. 2005. Multiple imputation of missing values: Update of ice.
Stata Journal 5(4): 527-536.
You can find these articles here:
http://www.stata-journal.com/archives.html
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
Yahoo! For Good helps you make a difference
http://uk.promotions.yahoo.com/forgood/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
