Date: Thu, 20 Sep 2001 07:36:58 -0700
Reply-To: Yvette <hollied@OLIN.WUSTL.EDU>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Yvette <hollied@OLIN.WUSTL.EDU>
Subject: Re: Times-series analysis in SAS
Content-Type: text/plain; charset=ISO-8859-1
vvgsgor@DE.STATE.AZ.US (Stanley A. Gorodenski) wrote in message news:<3BA8B7BE.E0DC129@de.state.az.us>...
> My suggestion is pretty simplistic, since I am not a sophisticated sas programmer. You probably have already thought of this.
> Sort in descending order by your time variable, create the lagged variables you want with the lag statements, and then match back
> to your original file by your time variable. I am sure there is a more efficient way to do this, but this is my 2 cents
> Yvette wrote:
> > I am trying to look at some data going forward rather than coming from
> > behind but I am not quite sure how to program this in SAS. Any
> > assistance would be appreciated.
> > For example, normally we might see the following equation:
> > Y(at time t) = X(t-1) + X(t-2) +X(t-3) + X(t-p) + some error term
> > HOWEVER, I am trying to examine the residual affects going forward
> > using a time-series analysis:
> > Y(at time) = X(t+1) + X(t+2) +X(t+3) + X(t+p) + some error term
> > I expect most of the residual affect to be found in the X(t+1) period
> > and then some beyond that but I'm not sure what P might equal at this
> > point.
> > Thanks.