A Rule-Based Dynamic Decision-Making Stock Trading System Based on Quantum-Inspired Tabu Search Algorithm

Abstract

Heuristic methods or evolutionary algorithms (such as genetic algorithms and genetic programs) are common approaches applied in financial applications, such as trading systems. Determining the best time to buy or sell stocks in a stock market, and thereby maximizing profit with low risks, is an important issue in financial research. Recent studies have used trading rules based on technique analysis to address this problem. This method can determine trading times by analyzing the value of technical indicators. In other words, we can make trading rules by finding the trading value of technique indicators. An example of a trading rule would be, if one technical indicator's value achieves the setting value, then either buy or sell. A combination of trading rules would become a trading strategy. The process of making trading strategies can be formulated as a combinational optimization problem. In this paper, we propose a novel method for applying a trading system. First, the proposed method uses the quantum-inspired Tabu search algorithm to find the optimal composition and combination of trading strategies. Second, this method uses a sliding window to avoid the major problem of over-fitting. The experiment results of earning money show much better performance than other approaches, and the proposed method outperforms the buy and hold method (which is a benchmark in this field).