Date:         Fri, 26 Sep 2003 15:04:36 -0700
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: data simulation
Comments: To: Regi <regivm123@YAHOO.COM>
Content-type: text/plain; charset=US-ASCII

Regi <regivm123@YAHOO.COM> wrote:
> I need to create a data set consisting of 7 variables. I know the
> paramters of the variables like mean and variance (all normally
> distributed). A also know the correlation coefficients between these
> variables. Can somebody help me with a possible approach?

Sure.  You can do it in SAS.  You should be able to create the data
set in a DATA step, unless your correlation matrix is too messy.  In
that case, it could be easier to do it using SAS/IML.

I can't help you much more than that, though.  You simply didn't provide
enough information.  Do you want a 7-dimensional multivariate normal
distribution?  How complex is the covariance matrix?  How many
observations
did you plan to create?  What are you using this for?

Perhaps, if you write back to SAS-L (not to me personally) with
more details, someone here can help you more.  In the meantime,
look through the SAS-L archives and you'll find that people have
written code to generate multivariate normal observations before,
and you could use that code as a starting point...

HTH,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician