Hello all,
 
We have a multivariate missing data(outcome is a continuous measure and
missing occurs in both outcome as well as in covariates) and the it
follows the assumption of MNAR. We are trying to do the pattern mixture
model approach as suggested by Little (1993) in his JASA paper. But we
had some difficulty of doing that. I can briefy explain below our data
structure:
 
The data set consists of four sets of patterns:
 
 1. Non- missing outcome(continuous measure) with all the values of the
covariates are available 
 2. Non-missing outcome with one or more vaues of the covariates are
missing 
 3. Missing outcome with  all the values of the covariates are available

 4. Missing outcome with one or more vaues of the covariates are
missing. 
 
We modelled on non-missing outcome using the first two groups and
estimated the parameters and using these estimated parameters, we
predict  the missing outcome, but for only with all the values of the
covatiate. We cannot be able to find out the values for the missing
outcome with missed covariate values. 
 
In this type of situation, how to model this case using the data
patterns in 3 and 4. Is there any method available for the missing
covarites for imputation. 
 
Any suggestions.
 
Thanks in advance.
 
Bala
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
http://lists.utsouthwestern.edu/pipermail/impute/attachments/20041025/2964f7dc/attachment.htm
From von-hippel.1 <@t> osu.edu  Thu Oct 28 11:50:34 2004
From: von-hippel.1 <@t> osu.edu (Paul von Hippel)
Date: Sun Jun 26 08:25:02 2005
Subject: [Impute] overloading SAS PROC MI
Message-ID: <[email protected]>

An HTML attachment was scrubbed...
URL: 
http://lists.utsouthwestern.edu/pipermail/impute/attachments/20041028/9a6df394/attachment.htm
From malone <@t> alumni.duke.edu  Thu Oct 28 11:55:36 2004
From: malone <@t> alumni.duke.edu (Patrick S. Malone)
Date: Sun Jun 26 08:25:02 2005
Subject: [Impute] overloading SAS PROC MI
In-Reply-To: <[email protected]>
References: <[email protected]>
Message-ID: <[email protected]>

On Thu, 28 Oct 2004 12:50:34 -0400, Paul von Hippel <[email protected]>  
wrote:

> I have a data set comprising 20,000 cases. When the number of variables  
> is more
> than 100 or so, I find that SAS PROC MI returns
>
> ERROR: Invalid Operation
>
> Has anyone else run into this.

Yes.

> I am using SAS 8.2.

That's your problem.  The limitation (which different people run into  
around 50-70 variables usually) was fixed by SAS 9.1.

>
> Thanks!
> Paul von Hippel



-- 
Patrick S. Malone, Ph.D., Research Scientist
Duke University Center for Child and Family Policy
North Carolina, USA
http://www.duke.edu/~malone
http://www.pubpol.duke.edu/centers/child/

Reply via email to