Date:         Thu, 28 Jun 2007 21:50:53 -0700
Reply-To:     David L Cassell <davidlcassell@MSN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         David L Cassell <davidlcassell@MSN.COM>
Subject:      Re: Cluster membership
In-Reply-To:  <0C094BFB78FEBC42B1BF98C83F49DB84965F@TEEXB0VD.prod.travp.net>
Content-Type: text/plain; format=flowed

JWANG@TRAVELERS.COM wrote:
>My data set has , say three variables, x y z, I am doing cluster
>analysis on x, y
>
>
>
>proc cluster data=a method=twostage k=10 noprint outtree=tree;
>
>var  x y;
>
>ID id;
>
>run;
>
>
>
>
>
>proc tree graphics horizontal out=out nclusters=10;
>
>copy x y;
>
>id id;
>
>run;
>
>
>
>My questions
>
>
>
>(1)  How to  let the output data (out or tree) contains a copy of z;
>
>(2)  How do I know which observation belongs to which cluster?
>
>
>
>Though I have specified nclusters=10,  the output data (out or tree) has
>more than 10 levels of "CLUSTER" or "CLUSNAME"
>
>
>
>I am a little bit confused.

Me too.

If your data set has 3 variables X, Y, and Z, then where does ID come from?

If you know how to drag ID along with the ID statement, why don't you
want to do the same with Z instead?

Why are you choosing METHOD=TWOSTAGE ?

Why are you specifying NCLUSTERS=10 when that only affects the
*output* data set?

HTH,
David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330

_________________________________________________________________
Don’t miss your chance to WIN $10,000 and other great prizes from Microsoft
Office Live http://clk.atdmt.com/MRT/go/aub0540003042mrt/direct/01/