應用資料探勘技術於股票投資 - 以日月光為例

Applications of Data Mining Techniques for Stock Investment - A Case Study of ASE Inc.

劉宗博(Zong-Bo Liu)

指導教授 ： 巫木誠 ； 洪暉智

國立交通大學/管理學院/工業工程與管理系所/碩士(2018年)

摘要

大數據時代來臨，資料探勘(Data Mining)技術統合了資訊工程、統計分析以及應用領域的專業知識，將過去累積的大量資料經由演算法轉換成更有價值的資訊，也因此應用範圍越來越廣。常見應用如股市預測、天氣、醫療診斷、交通狀況等。過去在股市方面的研究，多數是以預測明日股價或是判別漲跌為主。本研究以日月光公司股價為例，定義新的預測類別「買點信號」，在未來60日內，股價漲幅超過10%。運用多種資料探勘演算法建立預測模型。並將預測結果應用於不同股票交易策略，計算經由本研究建議之交易方法能夠獲得多少投資報酬率。

關鍵字

資料探勘 ； 股市預測 ； 股票交易策略 ； 投資報酬率

並列摘要

With the raising of big data, new techniques inherited from the fields of computer science and statistics are adopted in the latest data mining algorithms. These data mining algorithms accomplished with professional domain knowledges are verified to enable to dig out more valuable information from the existing data sets. Hence, data mining algorithms become more and more popular and are widely applied in many different areas. For example, stock market, weather forecasting, medical diagnosis, traffic forecasting, etc. In the stock markets, most studies apply data mining algorithms to predict the stock prices in future or to predict the bull or bear market in future. In this study, we focus on Advanced Semiconductor Engineering (ASE) Inc. and apply several data mining algorithms to build the prediction models. Our goal is to predict whether the next trading day is a “buy-day” or not. A trading day is called a “buy-day” if the stock closing price of ASE Inc. rises over 10% in the coming 60-day trading period. We also consider three trading policies and the overall return on investment (ROI) are computed to valid our prediction models.

並列關鍵字

Data Mining ； Stock Prediction ； Trading Policy ； Return on Investment

參考文獻

Breiman, L. (2001). “Random forests.” Machine Learning, 45(1): 5-32.

Breiman, L., J. H. Friedman, R. A. Olshen, and C. G. Stone. (1984). “Classification and Regression Trees.” Wadsworth International Group, Belmont, California, USA.

Burges, C. J. (1998). “A tutorial on support vector machines for pattern recognition.” Data mining and knowledge discovery, 2(2), 121-167.

Chong, E., Han, C., & Park, F. C. (2017). “Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies.” Expert Systems With Applications, 83, 187–205.

Cocianu, C., & Grigoryan, H. (2016). “Machine learning techniques for stock market Prediction. A case study of OMV Petrom.” Economic Computation and Economic Cybernetics Studies and Research, 50, 63-82.