Analysis of Chaotic Behavior of Artificial Stock Market realized by the Multi-agent Systems based on the Co-evolutionary Genetic Programming

摘要

This report deals with the analysis of chaotic behavior of multi-agent systems based on the co-evolutionary Genetic Programming (GP) in the artificial stock markets. Assuming five types of agents, in which rational agents prefer forecast models (equations) or production rules to support their decision making, and irrational agents select decisions at random like a speculation. Piecewise linear function is assumed in the decision of buy/sell stock due to the restriction of trading funds, which is resemble to the neural network. Then, we examine the chaotic behavior of agents by using the approximate representation of the system by a neural network.