Information diffusion and its impacts on artificial stock markets in different learning process

Empirical studies have divergence on whether the false information can affect market in the long term or not. Theoretical model can only explain part of these empirical studies. In this paper, we incorporate information diffusion model into an agent-based artificial market and test if the difference of investment experience leads the long time impact of false information. Our result provides an explanation to empirical studies that if connected agents are rich in investment experiences, they can distinguish false information under the condition of heterogeneous information sources. Consequently, informed agents have a low probability to release false information in the long term.