University of Twente Student Theses
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.
Machine Learning for Long-Term Stock Market Outperformance : A Data-Driven Approach to Identifying High-Growth Investments
Miedendorp de Bie, T.J. (2025) Machine Learning for Long-Term Stock Market Outperformance : A Data-Driven Approach to Identifying High-Growth Investments.
|
PDF
1MB |
| Abstract: | Investors consistently seek strategies to outperform market benchmarks over the long term. Yet, emotional biases and short-term volatility hinder optimal decision-making. With the advancements in machine learning and Explainable AI (XAI), models can systematically analyze historical financial data of companies to identify key indicators that are associated with long-term stock market success. This research proposes a data-driven approach - leveraging historical stock fundamentals and performance metrics - to identify stocks with a high likelihood of achieving a yearly return of at least 20\% over multi-year horizons. By incorporating expert investing insights and financial theory into the feature selection, this study aims to bridge human intuition and algorithmic precision on a large scale. Moreover, emphasis is placed on model explainability through tools such as SHAP values and permutation importance. This enhances transparency and trust in the investment predictions. The dataset includes publicly listed companies across Asia, Europe, South and North America, with more focus on the Central Asian and All Region markets. A key contribution of this paper lies in demonstrating that profitability-related indicators - such as Profit/Loss, Earnings Per Share, and 3-Year Revenue Growth Rate - play an important role in identifying high-growth investment opportunities. Furthermore, this study provides a Proof of Concept for integrating feature transparency into financial models. Some of the models achieved a precision exceeding 75\% in both the Asian and broader regional markets, highlighting their practical utility. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Awards: | Best Presentation Award |
| Link to this item: | https://purl.utwente.nl/essays/107192 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page