Date:         Fri, 13 Nov 2009 08:45:27 -0800
Reply-To:     Selina <cnfengshuang@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Selina <cnfengshuang@GMAIL.COM>
Organization: http://groups.google.com
Subject:      Re: Rolling window autocorrelation
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

Hi everyone,

Thank you very much for all your replies.  Rolling autocorrelation is
useful when I need a time series of autocorrelation to check it's
relation with other time series variables.

I have found a macro for rolling window regressions.

http://wrds.wharton.upenn.edu/news/sideitem/user2007/rollingreg.pdf

I have edited it so that it can calculate rolling window
autocorrelations.

The problem that I have is: the macro allows for only equal length of
by groups, i.e. it only has the same starting and ending dates for all
ID groups.  However, in my data, the id groups are of different
length.  Could anyone help me with it?  Thank you so much.

Regards,
Selina