Using extended classifier system to forecast S&P futures based on contrary sentiment indicators

Abstract

This research demonstrates the accurate forecasting performance of extended classifier system (XCS) based. on contrary sentiment indicators in predicting S&P 500 futures. These indicators include volatility index, put-call ratio, and trading index. To prove that XCS based on sentiment indicators can fit the financial forecasting domain, the performance of XCS is compared with that of three trading strategies, including buy-and-hold, trend-following, and mean-reversion strategies over the same sample period. The simulation results showed that XCS based on contrary sentiment indicators possesses both forecasting accuracy and profits earning capability in the real world.