A Study of Predicting Stock Trends Using Machine Learning Methods: Evidence Based on Listed Financial Companies in Taiwan

• Main Authors: LIN,YUNG-JEN, 林詠任
• Other Authors: LIN, FENG-CHEN
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/227c7r

碩士 === 逢甲大學 === 商學專業碩士在職學位學程 === 107 === In recent years, it has become a topic of increasing interest in the field of financial technology globally. Therefore, the use of various technological applications in finance industry has surged, and cross-border integration of innovation in finance and technology has made financial services more efficient. The motivation of this paper research is to predict the trend of financial stocks in the financial market using machine learning. The main purpose of this paper is to present the impact of financial news on financial stocks. Further research focus on the sensitivity of stock prices to positive news, negative news and general financial news. Due to the prevalence of information, investors can get information from internet at anytime and anywhere. Therefore, this paper mainly discusses the impact of financial news on investment intention, investors’ willingness and stock prices. With the technology index, a proposed model implemented to predict the stock price changes by machine learning is established. The results of the study show that when the market releases bullish news, the increase of stock price is obvious, and the decline of stock price is relatively clear when the market releases bad news.