University of Twente Student Theses

Login
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.

Data-driven stock selection using Machine Learning for StockWatch

Mink, R.J. de (2025) Data-driven stock selection using Machine Learning for StockWatch.

[img] PDF
3MB
Abstract:StockWatch is a Dutch stock content startup that offers analyses, articles, and portfolios through a subscription model. Currently, its stock-selection process lacks a systematic approach and its stock portfolios are unable to outperform the benchmark. This thesis addresses the following research question: How can a data-driven model be developed to support StockWatch in systematically finding outperforming stocks? To answer this, a Machine Learning model was developed, using 19 stock-specific and 5 macroeconomic characteristics. This process started with an OLS model, which was subsequently enhanced by two Machine Learning models (Random Forest and Neural Networks). The best performing model was an ensemble model using an equal weighting between Random Forest and Neural Networks. This model’s decile 1 portfolio achieved a Sharpe ratio of 0.94 on an out-of-sample test, delivering more than 50% excess returns per unit of volatility, and produced 22.61% excess returns per year. A screening tool was built to practically implement this model for StockWatch and help users find outperforming stocks from the S&P 500, AEX, and EURO STOXX 50 Index. This thesis recommends launching a Big Data portfolio based on this model to provide additional value to StockWatch’s subscribers.
Item Type:Essay (Bachelor)
Clients:
StockWatch, Amsterdam, Netherlands
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:31 mathematics, 58 process technology, 83 economics
Programme:Industrial Engineering and Management BSc (56994)
Link to this item:https://purl.utwente.nl/essays/107035
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page