Genetic Network Programming for Artificial Stock Market

Combined with the idea of Austrian school of economics, this paper proposes a new multi-agent model for artificial stock market based on genetic network programming. It focuses on applying the GNP approach to emulate investment behavior of agents and evolve their trading rules. Simultaneously, this model enhances the heterogeneity of agents, and searches for an optimal combination of parameter values based on GA. Simulation results confirm the effectiveness of this GNP-ASM model through comparison with empirical statistics