Stock trading decision support system using a rule selector based on sliding window

Abstract

We propose a dynamic decision support system (DDSS) capable of determining a near-optimal rule-combination for each time interval (window). The system provides Buy, Hold and Sell signals from which profitable trading decisions can be made. In DDSS, an intelligent rule selector(GARS) based on genetic algorithms and a sliding window scheme is developed. Experimental results on Taiwan stock exchange weighted stock index (TSEWSI) show that DDSS outperforms its static counterpart as well as the simple buy-and-hold strategy.