A Dynamic Stock Trading System using GNQTS and RSI in the U.S. Stock Market

Abstract

Investing in stocks is one of the most common options of financial management, and many people use technical indicators to decide when to buy and sell stocks. This research proposes a novel method of utilizing the commonly used technical indicator, Relative Strength Index (RSI), to the extreme. First, we remove the restriction of the traditional RSI’s default parameters and expand all the possibilities in RSI to find the best strategy and to maximize profit. Second, we employ a modified metaheuristic algorithm, the global best-guide quantum-inspired tabu search algorithm with quantum not gate (GNQTS), to efficiently search for the optimized parameters for RSI. Third, our approach also applies a sliding window to flexibly adjust training periods and avoid over fitting at the same time. The experimental environment covers popular indices and companies in the United States stock market such as DJIA, AAPL, etc. By removing the restriction of RSI, the experiment result shows that GNQTS can find optimized parameters of RSI to get more profit than traditional RSI and buy-and-hold strategies.