[Rquantlib-devel] RQuantLib

[Dirk Eddelbuettel]

Leo, Christian, Bryan:

I had pleasant chats with all three of you during R/Finance regarding future
work on RQuantLib.  I'm game.  A couple of ideas:

  - One-on-one email is inefficient. I would love for all of this to take
    place on the somewhat dormant rquantlib-devel list hanging off the
    project.  Whoever wants to follow up should be able to subscribe
    there. If not, ping me.

  - I also CC'ed Khanh who has some dormant interest in RQL, and Shane and
    Josh who both helped with the dreaded Windows builds for CRAN.

  - I think I want to toy with using Rcpp modules for simpler easier
    interfaces.

  - We need a better data strucure for fixed income instruments. What Khanh
    did in his Google Summer of Code project works great, but is
    inefficient. The termstr package on CRAN has something better.  We should
    try to maybe use their scheme.  Any takers -- Christian, would this
    possibly be up your alley as I presume pensam uses QL for fixed income
    (just a guess, I could be wrong.

  - Bryan mentioned something rather exciting about decomposing process
    generators and pricers at the R level, maybe modules could here there
    too. 

  - I just rebuild RQuantLib 0.3.7 against the QuantLib 1.1 pre-release
    snapshot I pushed into Debian. No changes needed / forced upon us. Good.

  - This was just a quick braindump, I am between kids soccer games this
    afternoon but had meant to get this out for the last few days.

Let's put some life into RQuantLib. 

Cheers,  Dirk

-- 
Gauss once played himself in a zero-sum game and won $50.
                      -- #11 at http://www.gaussfacts.com