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: multiple imputation when all the independent variables have no missing values


From   Mandy fu <mandy.fu1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: multiple imputation when all the independent variables have no missing values
Date   Fri, 5 Mar 2010 14:18:26 -0500

Dear all,

I met some difficulty of conducting multiple imputation for ordinal
categorical variables. It would be great if someone could give me some
suggestion.

The data set I have includes :

 *450 observations totally
 *204 ordinal variables (after imputation these 204 items are used to
calculate 5 scales, which are the dependent variables in the following
regressions)  have missing values
 *   4  independent variables (dummy variables)                 no
missing values at all

I used --ice-- and included all the 208 variables in the variable list
to be imputed using ologit.

I have two questions:
(1) When the independent variables have NO missing values and only the
variables used to calculate depended variables have missing values, is
it appropriate to use multiple imputation? I doubt I was off the right
track from the very beginning.

(2) If the answer to (1) is yes, I cannot understand why the computer
works unbelievably slowly when conducting the MI. It took 6 hours for
the computer to finish only 4 cycles and there are 6 cycles  left (I
set cycle (10) m(5)).

Thanks for your time!

Sincerely,
Mandy Fu
*
*   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