Application research in stock market forecast with fuzzy cognitive maps
模糊认知图在股票市场预测中的应用研究

There is a number of process relying, self organization and evolution in the complex system, so the traditional modeling method is difficult. Fuzzy cognitive maps as a soft computing method, which combines the synergistic fuzzy logic and theories of neural networks provides a powerful tool for the complex system modeling. According to the characteristics of fuzzy cognitive maps, a genetic learning algorithm was proposed for the construction of the system fuzzy cognitive maps. It provided a method for the analysis and the forecast of complex system. Finally, the stock market data were used as the test data for the simulated test of analysis and prediction. The result shows that the forecast technology has short-term forecast ability.