Date:         Tue, 22 Feb 2005 13:58:44 -0800
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: Carry-over residuals (for a newbie)
In-Reply-To:  <1109101636.673913.80030@z14g2000cwz.googlegroups.com>
Content-type: text/plain; charset=US-ASCII

Jason <falciparum@HOTMAIL.COM> wrote:
> I have a question about analyzing a data set and including carry over
> residual in the analysis.  If I have a design like the following(don't
> worry about the actual design--it's just hypothetical):
>
> A B C
> C D B
> D C A
>
> Where columns are people, rows are periods, and letters are
treatments.
>  Can I assign an arbitrary letter 'O' above each column and then
> include the assigned residuals as factors in my model?
>
> O O O
> A B C
> C D B
> D C A
>
> So then carry over residuals would be:
>
> O O O
> A B C
> C D B
>
> So my SAS code would be?...
>
> proc glm;
> CLASS Row Col trt Residual;
> MODEL y= Row Col trt Residual;
> Random Col Resid;
> Run;

I'm sorry, but I don't see what your question is.  Could you
explain what you are trying to do, and why you want to add in
your 'extra' treatments and throw out what looks like some
well-planned treatments?  Why not leave that last row in?

Are you interested in dealing with carry-over, or residual
influences, in an experimental design?  If so, what is supposed
to be the point of the 'initial' treatments?

If you have random effects, why are you using PROC GLM,
instead of PROC MIXED?

Where does your variable 'residual' come from?  You haven't
explained what it means or how you derive it.

I'm lost.  Perhaps you could write back (to SAS-L, not to me
personally) and explain what you want to do, and why you are
choosing an approach like what you describe above.  The more
you help us, the more we can help you.

David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician