Abstract

The classical approach to the modelling of complex dynamics systems is to use mathematical techniques such as differential equations to approximate observed global behaviours. For some systems, this approach is inadequate. An artificial trading system is specified using simple rules to conduct trade between entities which are represented using adaptive computational paradgims. Evolution is used to develop the entities, which complete to gain wealth. The global dynamics produced by the interaction of the individual strategies can be studied, to gain an insight into the behaviour of real trading systems, and of complex dynamic systems in general. The system provides a test-bed for experimentation on adaptive computational paradigms.