Agent-based artificial financial market with evolutionary algorithm

In traditional financial studies, existing approaches are unable to address increasingly complex problems. In this paper, an artificial financial market is proposed, in accordance with the adaptation market hypothesis, using artificial intelligence algorithms. This market includes three types of agents with different investments and risk preferences, representing the heterogeneity of traders. Genetic network programming is combined with a state-action-reward-state-action (SARSA)(lambda) algorithm for designing the market to reflect the adaptation of technical agents. A pricing mechanism is taken into consideration, based on the auction mechanism of the Chinese securities market. The characteristics of price time series are analyzed to determine whether excessive volatility exists in four different markets. Explanations are provided for the corresponding financial phenomena considering the hypotheses under the proposed novel artificial financial market.