[R-sig-finance] R vs. S-PLUS vs. SAS

Patrick Burns patrick at burns-stat.com
Fri Dec 3 19:37:15 CET 2004


There may be some differences between SAS procedures, but
at least generally SAS does not require the whole data to be in
RAM.  Regression will take the data row by row and do an update
for the answer.

For many things SAS is remarkably fast as well.   Some of their
routines are decades old, when machines were very much slower
so it paid to put a lot of effort into efficient algorithms and code.

If SAS does exactly what you want, you'll get an answer quickly.
If you want something slightly different than what SAS does, you
are probably stuck.  That is one of the major differences between
SAS and R.

Patrick Burns

Burns Statistics
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Muller, John wrote:

>I have a question related to the discussion of R versus S-Plus versus
>SAS
>regarding handling large data sets.
>
>I too have often heard that SAS is better with large data sets
>but never sure whether "better" meant that the others had hard
>limits on data set size or simply that SAS is faster.
>
>Any insight into this from the group?
>
>For example, does SAS use special algorithms (e.g. regression,
>clustering,
>tree construction) that do not require all the data in RAM
>OR does SAS do very clever sampling OR ...?
>
>I think SAS's Enterprise Miner does sampling by default.
>
>Thanks for any insight you can offer.
>
>- john muller
>-------------------------------------------------
>John H. Muller
>mailto:john.muller at bankofamerica.com
>404.607.5943
>
>_______________________________________________
>R-sig-finance at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>
>
>  
>



More information about the R-sig-finance mailing list