Date:         Tue, 30 May 2006 22:56:46 -0700
Reply-To:     David L Cassell <davidlcassell@MSN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         David L Cassell <davidlcassell@MSN.COM>
Subject:      Re: Adapting/converting R-software programs to SAS: Any crosswalk
In-Reply-To:  <200605301110.k4UAlvtf005689@malibu.cc.uga.edu>
Content-Type: text/plain; format=flowed

yungjui@GMAIL.COM replied:
>Many crosswalks do exist when I translated John Fox's R polyserial
>function into SAS.  But some don't and I take pains to develop them
>while translating all the codes.  For example, the pmvnorm() in R is
>the density distribution function of multivariate normality, but none
>is available in SAS, and one has to rely on integration to generate
>such function! Besides, "-inf" and "inf" are not allowed in a SAS/IML
>matrix.

But those crosswalks occur because you stuck strictly to SAS/IML.

If you had taken Fox's code and translated it into code using
PROC NLMIXED or PROC MODEL (assuming that is even feasible),
then there would have been no crosswalk, because the key features
would be addressed in totally different ways.  That was the point
I was trying to make before.

Just look at bootstrapping a variance estimate for a complicated mixed
model in R, and the same in SAS (without doing it in IML!).  You'll see
very different code, stressing very different aspects of the two
languages.

David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330

_________________________________________________________________
On the road to retirement? Check out MSN Life Events for advice on how to
get there! http://lifeevents.msn.com/category.aspx?cid=Retirement