On optimality condition of interacting agents in financial markets

The promising approach to understand financial markets through interactions of agents is limited by the major unresolved problems of complex systems. In particular, the laws and mechanisms that translate interactions of agents into their desired collective behaviour are still unknown. Optimality condition when interacting agents as a whole show they [sic] best performance for a particular problem is sought in the paper. Results of extensive computational experiments give strong facts to believe that such an optimality condition exists and can be formulated in terms of a concept of structural complexity. Namely, the agents show their best performance for a problem when their structural complexity equals the structural complexity of the problem.