Empirical Research On Traders' Personal Evolution Mechanism Based On CSI 300 Index

The traditional classical financial theory formed its core content——"efficient market hypothesis", on the basis of " rational man" hypothesis。According to this hypothesis,all kinds of financial theory rose, including the portfolio theory, the capital asset pricing model, arbitrage pricing model and option pricing model ,and constructed the traditional classical financial theory foundation and these theoretical models of the current main constituted securities investment strategy theoretical basis. However, in real financial securities market activity , there are many contrary to traditional classical financial theory "vision", such as, stock long-term investment rate of return, stock price premium of abnormal fluctuations and share price bubble, the mystery of the closed-end fund, share the information of the market reaction or excessive reaction shortage, and so on. These vision challenge to the traditional classical financial theory. One of the focus of debate is the strict hypothesis. numerous research results show that it is difficult to reach the market participants completely rational degree, and showed a significant limited rational characteristic.In this paper, the hypothesis is the market participants are heterogeneous limited rational , simulate the decision rules of traders through the neural network model ,and establish genetic algorithm model to simulate the process of evolution traders learning, then observe the study of evolution action traders how to influence stock market from down to up.The results of the study showed that under different learning speed, the volatility of the stock market showed different characteristics. We make CSI 300 index represent the market, to empirically test the stock market fluctuation in different learning speed. Regression testing results show that the stock market volatility in influence of the rate of turnover on behalf of learning speed to is remarkable. Simulation and empirical results show that the, the learning speed of traders has significant effect on the stock market.