Re: Input on choice of programing language

[Nick Collier]

You might also want to take a look at the evo toolkit. It is written on
top of Swarm. It may or may not be what you are looking for. See,

http://omicrongroup.org/evo/

for more info.

Nick

On Thu, 2002-10-17 at 09:58, Marcus G. Daniels wrote:
> >
> >
> >I want to build an artificial
> >market - in contrast to the SFI-ASM not with GAs but with genetic
> >programming - and to assess the impact of different regulations and auction 
> >systems on the market effiency. For rigorous analysis, I will need quite 
> >powerful statistical tools to calculate heteroskadicity, leptokurtusis etc. 
> >I am currently pondering, what programming language I should choose.
> >
> Take a look at R.  R is a Lisp-like language and has extremely powerful 
> statistical functions.
> Being Lisp-like, you can create expressions dynamically and evalulate 
> them, as is needed in GP.
> There is also a C interface to R features in case you need to make 
> blocks of simulation code that run natively.  In other words, you can 
> create dynamic agent behaviors from C (or Swarm) if you need to for 
> performance.   Since most of GP is fitness evaluation (of interpreted 
> code), it may well be the big bottleneck in the simulation, and so you 
> might be just as well of writing the whole thing in R.
> 
> 
> 
>                   ==================================
>    Swarm-Modelling is for discussion of Simulation and Modelling techniques
>    esp. using Swarm.  For list administration needs (esp. [un]subscribing),
>    please send a message to <address@hidden> with "help" in the
>    body of the message.
>                   ==================================
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net



                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================