Date:         Wed, 10 Apr 2002 08:55:03 -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: from IML to data set
Content-type: text/plain; charset=us-ascii

Carlos Tadeu <ctadeu@NOETHER.EX.AC.UK> wrote [in part]:
> I have the following code in IML and I would like:
> 1) put the variables S G R and Y in a data set (MYSGRY).
> Note that they are of different size, but I intend put toghether
repeating S, G and R. So my problem is how to put this
> variable into a unique file.
> 2) I would like to run PROC GLM with that data set one hundred times
and get the estimates
> of parameters into another data set for posterior analysis. So my
problems are: how can I get
> this parameters and how can I run PROC GLM into proc IML.

It looks to me as if you should consider programming this in a DATA
step instead of IML.  Your code does not show anything that is IML
specific, just what I assume to be some private functions you did
not bother to show us.  Then you could use data step arrays to do
the math, output all the desired lines, and run PROC GLM using the
BY statement [BY ITER;] to get all the iterations run separately.
Then you would have the results in an output data set.

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