Tim Chenoweth – A Multi-Component Nonlinear Prediction System for the S&P 500

The proposed stock market prediction system is comprised of two preprocessing components, two specialized neural networks, and a decision rule base. First, the preprocessing components determine the most relevant features for stock market prediction, remove the noise, and separate the remaining patterns into two disjoint sets. Next, the two neural networks predict the market’s rate of return, with one network trained to recognize positive and the other negative returns. Finally, the decision rule base takes both return predictions and determines a buy/sell recommendation.

Daily and monthly experiments are conducted and performance measured by computing the annual rate of return and the return per trade. Comparison of the results achieved by the dual neural network system to that of the single neural network shows that the dual neural network system gives much larger returns with fewer trades. In addition, dual neural network experiments with the appropriately selected filtering and decision thresholds managed to achieve an almost twice larger annual rate of return when compared to that of the buy and hold strategy over a seventy month period. However, no claims are made that the proposed system is better than the buy and hold strategy when considering transaction costs.