Machine learning for value investing

• Main Author: Jin, Zhenyu
• Other Authors: Wang Lipo
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/173598
• Institution: Nanyang Technological University

With the gradual elimination of the impact of the epidemic, financial markets have also shown a gradual recovery trend. At this time, how to obtain a higher return on investment under the premise of low risk has become an important knowledge. As a classic investment strategy, value investment relies on in-depth analysis of the company's fundamentals, which has lower risks than other investment strategies. And machine learning has unique advantages in data processing. This study thus proposes an innovative machine learning-based value investing model. Through the analysis of the annual financial data (revenue, profit, assets, liabilities and other financial indicators) of thousands of listed companies by using ridge regression, aim to predict the future stock price of enterprises. Through in-depth analysis of ten years of financial data of these companies, we not only verify the correlation between financial characteristics and future stock prices, but also confirm the feasibility of machine learning models in the field of value investing. In the end, we filtered out some companies that are likely to achieve higher returns in the future, providing a strong reference for investors. This study can be seem as a successful and effective attempt in the intersection of value investment and machine learning, which provides a solid theoretical support for the subsequent in-depth exploration of value investment and financial decision-making.