They are two totally different algorithms for obtaining clusters. PROC FASTCLUS uses the k-means algorithm. It is an iterative partitioning algorithm instead of a hierarchical agglomerative algorithm so you would not get the same kind of output.

Dennis G. Fisher, Ph.D. Professor and Director Center for Behavioral Research and Services California State University, Long Beach 1090 Atlantic Avenue Long Beach, CA 90813 tel: 562-495-2330 x121 fax: 562-983-1421 http://www.csulb.edu/centers/cbrs

-----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of OR Stats Sent: Thursday, June 18, 2009 12:31 PM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: PROC CLUSTER... slow

Thanks!

When I tried using PROC FASTCLUS, it does not return

Eigenvalues of the Covariance Matrix that PROC CLUSTER does, which gives me a good idea of the # of vectors of my correlated covariates that explains most (>=80%) of my total covariates variance. How do these two procedures differ in computational algorithm?

On Thu, Jun 18, 2009 at 1:00 PM, Dennis Fisher <dfisher@csulb.edu> wrote:

> Usually when you have that many observations you would use PROC FASTCLUS > instead. > > Dennis G. Fisher, Ph.D. > Professor and Director > Center for Behavioral Research and Services > California State University, Long Beach > 1090 Atlantic Avenue > Long Beach, CA 90813 > tel: 562-495-2330 x121 > fax: 562-983-1421 > http://www.csulb.edu/centers/cbrs > > -----Original Message----- > From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of OR > Stats > Sent: Thursday, June 18, 2009 10:40 AM > To: SAS-L@LISTSERV.UGA.EDU > Subject: PROC CLUSTER... slow > > Hello: I have some 2,000 records for which I am running PROC CLUSTER on 7 > variables. After almost a full 24h, it still was not done. I am running > XP > with E5440 @ 2.83 GHz 2.83 GHz at 3.25 GB of RAM. Does anyone know why it > is so slow? Thx so much!! > >