Day Trading Strategies in Taiwan Stock Index Future by Using Evolutionary Algorithm and Technical Indicators

• Main Authors: Ku, Kai-Yun, 顧凱云
• Other Authors: Chen, Ying-Ping
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/4tw3vg

碩士 === 國立交通大學 === 資訊科學與工程研究所 === 106 === Technical analysis has been practiced in the stock market for years. In this study, we take the Taiwan stock index futures as the research object. Refer to previous research[1], we use the Stochastic Oscillator (KD), exponential moving average (EMA) and the volume (MV) index to find the suitable time to buy the future and sell it according to the stop-loss point and lock-in-gain point by using the evolutionary algorithm. We use historical data from November 2012 to December 2016 to develop day trading strategies. Our trading strategies use two months historical data to run evolutionary algorithm (training) and use the next month historical data to simulate trading by using optimized parameters provided by evolutionary algorithm (testing). This mechanism will shift one month and continue executing until December 2016. The result of this study show that there are positive profits from year 2013 to 2016.