|Date: ||Thu, 2 Sep 2004 12:38:31 -0400|
|Reply-To: ||Andrew Bolton <Andrew.Bolton@IDENTEX.COM>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||Andrew Bolton <Andrew.Bolton@IDENTEX.COM>|
|Subject: ||Re: ot: stats question|
Hi Paul, Jim,
So using one ordinal variable (0,1,2 etc) in the clustering is OK (ish),
but there may be issues associated with the band width (if they are
uneven) and relative frequncies within the band (if there are too few or
too many bands)?
Do you have any references I could look at for this? If so I'll go and
have a dig about.