Self-organizing neural network system for trading common stocks

A fully automatic common stock trading system has been developed. The system takes in daily price and volume data on a list of 200 stocks and 10 market indexes. A chaos based modeling procedure is then used to construct alternate price prediction models based on technical, adaptive, and statistical models. A self-organizing neural network is used to select the best model for each stock or index on a daily basis. A second self-organizing network is then used to to make a short-term gain-lose prediction from each model. These predictions are combined in a trade selection module to generate buy-sell-hold recommendations for the entire list of stocks on a daily basis. Finally, the trading recommendations are combined by a portfolio management utility to produce a set of risk-reward ranked alternate portfolios.<>