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 on April 23, 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


From   weddings@stata.com (Wesley D. Eddings, StataCorp)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: multiple imputation
Date   Thu, 02 Dec 2010 16:27:36 -0600

Cecilia <xqcecilia@msn.com> received from -mi impute regress- an error message
about missing values:

> I am using  mi in stata to impute missing data for panel data. But STATA gives
> me the error reminder "This may occur when imputation variables are used as
> independent variables or when independent variables contain missing values.
> You can specify option force if you wish to proceed anyway." My code is "mi
> impute regress loggdp_N x1 x2 x3 x4 x5 x6,add(20) rseed(5)". Is that because
> there are missing values in x1~x6 variable list which are used in the
> regression model to impute?

Yes.  The quoted error message is usually caused by missing values in the -mi
impute- predictor variables, and Cecilia may use -misstable summarize- to check
x1-x6 for missing values.  If the variables have no missing values, Cecilia may
send her data to Stata Technical Support at tech-support@stata.com for us to
diagnose a reason for missing imputed values.

Cecilia also asked about the -force- option:

> For this error, I add "force" in the command, but no imputation generated at
> all.

We aren't sure what Cecilia means by the imputations not being "generated." The
-mi impute regress, force- command should have added twenty imputations to the
data; if it didn't, we'd like Cecilia to send her data and do-file to Stata
Technical Support.

But Cecilia may mean that the imputations were added, but all of the imputed
values of "loggdp_N" turned out to be missing.  It is possible for all the
imputed values to be legitimately missing, if every observation with a missing
"loggdp_N" value was also missing one of the predictors.  Cecilia would need to
consider imputing all of the variables simultaneously using one of the
multivariate imputation methods.

  
-- Wes                  
weddings@stata.com      
*
*   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