Re: Are there theorems for Agent Based Models

[glen e. p. ropella]

At 03:27 PM 7/12/99 -0500, you wrote:
>An economist said to me the other day, "where are your theorems? Aren't
>there established results that guide modeling choices?"  I shrugged.

Just to throw a little spice into this rather pragmatic discussion,
before we can get to the point where "we" (by we  I 
mean ABMs) can even make *statements*, much less prove
theorems, we need an alphabet and a grammar. 

The closest I've seen to this kind of formality is in some of the 
distributed computing circles where they have lexica (lexicons?)
that are well-defined enough to tolerate attempts at isomorphisms
between computing architectures and things like Petri nets or 
partial orders.

I believe (please correct me if I'm wrong) that the kinds of results
cited by Sven are not really formal statements (theorems) so much
as they are "natural laws" induced from experiments.  These can
be formalized, of course, which they probably are.  But, they are more
akin to the mathematical modeling one finds in disciplines like control
theory or CFD than they are to the discrete math most often prevalent
in computing sciences.

We need both, of course. [grin]

glen

--
glen e. p. ropella =><= Feeding the hamster wheel.  Hail Eris!
Home: http://www.trail.com/~gepr/home.html      (505) 424-0448
Work: http://www.swarm.com                      (505) 995-0818


                  ==================================
   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.
                  ==================================