Simulation Of Stock Market Based On Cellular Automata

Influenced by the developments in the financial market and the theory of complex systems, people found that Efficient Market Hypothesis (EMH) is not valid. Stock returns are unstable stochastic series, the distribution of stock returns are not Gaussian distribution, stock returns exhibits character of non-linear and autocorrelations. More and more economists and physicists have been attracted to find price mechanism, and established a lot of models by complex systems tools.In recent years, Cellular Automata (CA) has been used widely to study complex systems. In this paper, we establish the stock market based on Cellular Automata with the character of rapid operation and simple evolvement rules of CA.The stock market environment is usually much more ill-defined and makes it impossible for traders to form precise and objective price objective expectations. To model the uncertainty of the market as well as how traders react when they face the uncertainty market, this paper we explore the use of an inexact algorithm for probability reasoning for dealing with the uncertainties involved in the market, in contrast to the stochastic models with deterministic parameters. We divide traders participate in the trading into three group and each group have their own "definition" of price changes degree, but the "definition" is fuzzy. Groups make decision to buy or sell stock in accordance with fuzzy "definition". So by introducing a stochastic cellular automata model with fuzzy parameters to simulate the dynamics of the stock market, some of well-known stylized facts of the stock market, such as fat tails, absence of autocorrelations, volatility clustering and multifractal proprieties are reproduced. A direct comparison is made with the daily closures of the Hang Seng composite index. We also study the market stability under different proportion of traders.