Economic Experiment with Swarm:

A neural network Approach to Self-Development of Consistency in Agents' Behavior

 

By

Pietro Terna


It investigates with much sensibility, the methodological implications of agent-based models for social sciences in general and economics in particular.

Methodology and techniques are ``largely under construction'' according to Terna, who proceeds to present a generalized Environment-Rules-Agent scheme as a tool to assist the researcher in the design of a ``bottom-up'' model.
A package bp-ct is proposed as an easy way in building and running artificial laboratories for social scientists. The package is finally employed to produce a spontaneous no-auctioneer Hayekian market.


Press here to download the source