[Maxima] [Maxima-users] Statistical Analysis.
blindglobe at gmail.com
Wed Jan 16 16:22:59 CST 2008
On Jan 16, 2008 5:04 PM, Robert Dodier <robert.dodier at gmail.com> wrote:
> On 1/15/08, A.J. Rossini <blindglobe at gmail.com> wrote:
> > Which reminds me to ask -- where do I look to see how to integrate
> > Common Lisp packages with Maxima?
> There are some notes in the reference manual under the heading
> "Lisp and Maxima", and also in a tutorial named "Minimal Maxima".
> You can find both documents at: http://maxima.sourceforge.net/docs.shtml
> See also these pages at the Maxima wiki:
> Sorry that this stuff isn't better organized. I guess we should merge
> it together ....
This is exactly what I'm looking for at this point. To answer Ray's
question in a round-about indirect way, the first use case I have is
for ODE and integral simplification, i.e. I'd like to be able to send
a CL form specifying one or the other, and get a simplified form back.
I'm assuming that I might need to appropriately translate or embed,
but in a sense, I'd like to be able to do something like write a
MaximaWrapper package which I could then do:
(setf OldFunction '(form denoting in lisp notation (some expression)
to (simplify) which would be ODE, integral, or mixed))
(setf newFunction (Tony'sMaximaWrapper::simplify OldFunction)
for later eval'ing of the newFunction in the right
closure/environment. Basically, stripping away the current Maxima
notation into something more directly CL-like.
Robert hit it on the head elsewhere in this thread -- one statistical
application I'm quite keen on is Bayesian nonlinear heirarchical
regression analysis where the mean form is ODE specified. (kinetics
equations in systems biology and pharmacokinetics).
> > (3 reasons -- I've been working with a Common Lisp <-> R gateway,
> > mostly R called from Common Lisp; I'm slowly working on a port/update
> > of LispStat from XLisp to Common Lisp, and I've got a strong interest
> > in leveraging Maxima from the CommonLispStat code base at some point,
> > probably next year given my other commitments).
> What is the mechanism for communicating between CL and R ?
> Is it a foreign function interface or a pipe or what?
Built on CFFI.
(there's another one out there built directly using SBCL's FFI as well).
blindglobe at gmail.com
"Commit early,commit often, and commit in a repository from which we
can easily roll-back your mistakes" (AJR, 4Jan05).
Drink Coffee: Do stupid things faster with more energy!
More information about the Maxima