Date:         Fri, 29 Sep 2006 13:25:53 -0700
Reply-To:     Reeza <fkhurshed@HOTMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Reeza <fkhurshed@HOTMAIL.COM>
Organization: http://groups.google.com
Subject:      Outlier Detection in Time Series Data
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset="iso-8859-1"

Hi All,

I'm more familiar with GLM's and OLS regression than I am with time
series, but I need to run an AR(2) model in SAS, and do so using proc
autoreg.
For OLS there are several methods to test the influence of a particular
observation, and output them in SAS.

so first of all, do the same statistics apply to time series, ie Cooks
D?
and how may I implement it in SAS?  Is there an  option or do I need to
calculate it?

Thanks,
Reeza