Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: micombine


From   Elisabetta Petracci <elisabetta.petracci@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: micombine
Date   Tue, 11 Jan 2011 17:38:46 -0500

*Thanks very much for your suggestions.
*Yes, I noticed the presence of a lot of commands new and old for
multiple imputation.
*My choice is on "mi" in Stata 11. See the follwing steps:

use http://www.stata-press.com/data/r11/mheart5 (1)
mi set mlong (2)
mi register imputed age bmi (3)
set seed 29390 (4)
mi impute mvn age bmi = attack smokes hsgrad female, add(10) (5)
mi estimate: logistic attack smokes age bmi hsgrad female (6)

*Suppose that missing values were on the dependent variable "attack"
(Y) and my interest is not in the logistic estimation but just in
*obtaining a complete Y.

*Assuming m=10, what about taking the mean of each missing and then
imputed observation for Y? Does it make sense or anyone *has another
suggestion?

Thanks,
Elisabetta




2011/1/11 Maarten buis <maartenbuis@yahoo.co.uk>:
> --- On Tue, 11/1/11, Elisabetta Petracci wrote:
>> I am using Stata 11 and referring to the paper published
>> on Stata Journal "Multiple imputation of missing values"
>> by Royston (2004). Following the logic of the reported
>> example, I need to impute missing values for dichothomous
>> varaibles and then to run  a quantile regression on the
>> complete datasets, considering m=5 copies of the dataset.
>>
>> "micombine" should work with command "qreg" but I get the
>> following
>> error message: "imputation identifier _mj not found"
>>
>> Why does this happen? Any suggestion to impute missing
>> values using a Multiple Imputation technique able to run
>> a quantile regression afterward?
>
> There have been many follow up articles on this since. The
> consequence is that there are now many versions of both
> -mvis-, -ice-, -micombine-, and -mim- floating around in
> cyberspace. If your commands are not in sync, or if you use
> syntax appropriate for another version than the one on your
> system, then you can easily get this type of error message.
> Safest is to go for the latest version of -ice- (the
> successor of -mvis-) and -mim- (the successor of
> -micombine-). Both have been published in the Stata Journal,
> the articles can be found by typing in Stata: -findit ice-
> and -findit mim-. The latest versions of the software is
> available from SSC, type in Stata -ssc install ice- and
> -ssc install mim-.
>
> 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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index