Model and simulation of stock market based on agent

Stock market is an extremely complex nonlinear dynamic system. Neural network has the capability of approximating any nonlinear system and the specialty of self-learning and self-adapting, attempt to substitute neural network for agent. Secondly, refer to the modelling scheme of Hollandpsilas stock market model, an applied model of stock market using ERA scheme was built. Finally, the dynamic evolvement of the stock market under uncertain environment was simulated through the neural network approach to the self-development of consistency in agent behaviour with CT method. Aim to understand the dynamic specialty of stock market deeply, instead of try to predict.