The stylized properties of artificial stock market comparison with Chinese empirical data

In the decade,the use of agent-based simulations of market has been acceptance by more and more social scientists. This paper presents the construction of the artificial stock market that the artificial intelligent algorithms take on the role of trades. They make predictions about the future, and buy and sell stock as indicated by their expectations of future risk and return. Prices are set endogenously to clear the market. Time series from this market are analyzed from the standpoint of well-known empirical features in Chinese stock market. The artificial stock market is able to replicate several of these phenomenon, including volatility clustering, the excess kurtosis of the distribution of return, especially fractal structure. From this point, the artificial stock market can not only generate stock price trends and properties rather similar to the real stock market, but also show the fractal structure in deep consistency with the real stock market. So, research on the artificial stock market can reveal evolution rule of real stock market, operate mechanism, policy influence and better investment strategy.