[Swarm-Modelling] [Fwd: Re: [ABMs in finance]

[Steve Railsback]

-------- Original Message --------
Subject: Re: [Swarmfest2006] Keynote speakers
Date: Fri, 21 Apr 2006 07:52:23 -0600
From: Marcus G. Daniels <address@hidden>

Pietro Terna wrote:
>          But how can you obtain results like that of [examples removed]
To bring the topic back to the purpose of this list, and, say, having
Doyne speak on behalf of ABM or not:  His recent work uses very simple
dynamical models with little or no internal state.   That's because
prices are not driven by the diversity of elaborate strategies, but
rather what they have in common:  greed, fear, and the market mechanism
(the orderbook).  Implementing many paranoid greedy agents (hiding their
extent of their inventories, for example), doesn't really require ABM in
the sense of heavyweight Swarm-like models.  Drawing from probability
distributions is enough.    So I think in some sense if Doyne were to
speak he'd provide a skeptical view for appropriate use of ABM (which
might  actually be good).    Note the title of the Farmer/Patelli/Zovko
paper:  Zero intelligence.  That wasn't ABM work..