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: Re: st: remaining missings after multiple imputation


From   "Ploutz-Snyder, Robert (JSC-SK)[USRA]" <robert.ploutz-snyder-1@nasa.gov>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: Re: st: remaining missings after multiple imputation
Date   Tue, 20 Apr 2010 10:06:04 -0500

Thanks Maarten,
Indeed, it's a bit frustrating with user-written commands sometimes, but I think ice is doing a better job than mi for me because of my requirements to have specific user-identified imputation equations.  Plus there's more in the literature to draw from on ice than mi, given that mi is currently on its maiden voyage with release of Stata 11.

The missing imputed values I spoke of were a result of uvis (not ice) and uvis just writes the newly imputed variable in the working dataset under a user-specified name.  That much I am sure of.  

I'm not asking for more help at this point--just following the thread.
Rob


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten buis
Sent: Tuesday, April 20, 2010 9:40 AM
To: statalist@hsphsun2.harvard.edu
Subject: RE: Re: st: remaining missings after multiple imputation

--- On Tue, 20/4/10, Ploutz-Snyder, Robert wrote:
> I'm getting similar results with some imputations that I'm
> working on, but I'm using ice instead of mi.  Strictly
> speaking, I'm working with uvis prior to ice to fine-tune my
> modeling of the missing values, and  I get missing
> imputations sometimes that I cannot explain.  

This is also hard to diagnose, and you can't contact Stata's
tech support for this, as -ice- is a user written program. 
So, you need to look for some pattern in your data that 
makes the model unidentified for particular observations.
However, before you do that, did you check that you only 
looked at the observations where _mj > 0 (as -ice- leaves
a copy of the original data, with missing values, behind
as _mj==0)?

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/

*
*   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