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]

st: Using mifit after ice to get imputed values


From   Ilke UNDER <onurunder@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Using mifit after ice to get imputed values
Date   Fri, 16 Jul 2010 09:32:24 -0600

Dear all,

I am using Stata 10 to do multiple imputations with ice and want to
draw point estimates combining the imputations created by ice. In a
previous email (though it is 2006), it is advised to use uvis, miset
and mifit to get the imputed values for the initial data set.
Currently, I assume ice is more efficient since it also allows using
predictors with missing values and shortens the overall procedure.
But, I had problems using misplit (update for miset) followed by mifit
to get the final imputed values combining the multiple imputations.

Do you think mifit is the appropriate tool for my purposes (or
micombine applicable?)  Carlin, 2003; Royston 2004, 2005 does not make
it clear for me.

Here is an example showing my problem:

ice PS_INCOME PS_FEMALE PS_EMPLOYM PS_EDU PS_MARI PS_EDU4 PSAGE,
saving (IMP_income) boot cmd(PS_INCOME: ologit) m(5)

use IMP_income, clear
rename _mj _j
misplit, clear

mifit: ologit PS_INCOME PS_FEMALE PSAGE PS_EMPLOYM PS_EDU PS_MARI PS_EDU4

* I tried other variables using regression rather than ologit; mifit
did not work with them either.

Thanks for your interest,
*
*   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