You can do this with proc mianalyze, I believe.

You need to use ODS to save the parameters (and the covb matrix,
depending on what tests you want) and then you can use MIanalyze.

Jeremy

2009/11/11 David Judkins <[email protected]>:
> Well, I think this is the first question to the group since list ownership
> changed.  I wonder how many people are signed up now?  It hasn’t been a very
> active list for a long time.  Anyway, here is my question.
>
>
>
> I have a dataset with multiple imputations.  It is from a five-arm GRT.  One
> arm is a control and the other four are active.  I want to test for
> variation in mean responses across the four active arms.  Proc Mixed will
> give me a test statistic based on each multiple imputation.  But how do I
> combine these?
>
>
>
> One of colleagues found something in the HLM manual that would suggest that
> the replicates of test statistics other than t-statistics are averaged with
> no attention paid to the variability among them.    Sound accurate about
> HLM?  Is that the best we can do?
>
>
>
>
>
>
>
> David Judkins
> Senior Statistician
> Westat
> 1650 Research Boulevard
> Rockville, MD 20850
> (301) 315-5970
> [email protected]



-- 
Jeremy Miles
Psychology Research Methods Wiki: www.researchmethodsinpsychology.com

Reply via email to