Date: Wed, 8 Sep 2010 22:39:34 +0800
Reply-To: Murphy Choy <goladin@GMAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Murphy Choy <goladin@GMAIL.COM>
Subject: Re: first order autocovariance series
Content-Type: text/plain; charset=ISO-8859-1
I believe you are referring to auto correlation function? If so, you can try
PROC ARIMA with identify keyword.
On Wed, Sep 8, 2010 at 10:24 PM, MM <WBS.PhD@gmail.com> wrote:
> I am aware the question may be not difficult, but I still need some help.
> To simply, say my dataset has only to variables:
> DATE RET
> where RET=daily return on a particular stock
> I need to construct the time series of the first order autocovariance of
> returns.That is, for each observation (day) I need the covariance btw the
> return corresponding to that observation and the return in the previous
> observation (day before).
> Please suggest me a more efficient procedure than the one I am about to
> write down.
Certified Advanced Programmer for SAS V9
Certified Basic Programmer for SAS V9
DataShaping Certified SAS Professional