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

[Joe Conway]

Andrew Piskorski wrote:
> Someone might want to ask Joe Conway about his experience and thoughts
> integrating R as a procedural language inside PostgreSQL, to create
> PL/R:
> 
>   http://www.joeconway.com/plr/
>   http://gborg.postgresql.org/project/plr/projdisplay.php
> 
> (Hm, for good measure, I have Cc'd him on this email.)  Obviously, an
> RDBMS like PostgreSQL is expert at dealing with data that doesn't fit
> into RAM.  I've no idea whether PL/R does anything special to take
> advantage of that, or how feasible it would be to do so.

There is nothing particularly special in PL/R per se, but there are some 
advantages of the RDBMS-embedded R-interpreter approach of PL/R. 
Although not written about PL/R, much of Duncan Temple Lang's whitepaper 
related to usage scenarios for his REmbeddedPostgres package applies. It 
can be found here:

   http://www.omegahat.org/RSPostgres/Scenarios.pdf

Additionally, since PL/R supports direct in-process SQL queries from R, 
large data sets can be processed in groups of records (e.g. by using 
"DECLARE ... CURSOR FOR SELECT ..." and "FETCH 1000 FROM ..." in a loop) 
as long as the analysis being performed can support it.

I got the impression somewhere that the DBI packages support a kind of 
virtual data.frame interface to supported databases -- i.e. the rows are 
retrieved in groups as they are accessed, in some sort of FIFO manner. 
Is that true? Perhaps at some point that notion could be added to PL/R 
as well (no promises on timing, howver -- my day-job is keeping me 
pretty busy these days).

Joe