Publicação
EURIBOR Fluctuations and Property Values: A Machine Learning Comparison of Portugal and Spain
| datacite.subject.fos | Ciências Naturais::Ciências da Computação e da Informação | pt_PT |
| dc.contributor.advisor | Neves, Maria de Fátima dos Santos Trindade | |
| dc.contributor.author | Moita, Lara Laranjeira | |
| dc.date.accessioned | 2025-06-25T14:23:01Z | |
| dc.date.embargo | 2027-06-24 | |
| dc.date.issued | 2025-06-24 | |
| dc.description | Dissertation presented as the partial requirement for obtaining a Master's degree in Data Driven Marketing, specialization in Data Science for Marketing | pt_PT |
| dc.description.abstract | This study explores the predictive relationship between macroeconomic, demographic, and financial variables and the 6-month EURIBOR rate, examining how its fluctuations shape housing market dynamics in Portugal and Spain. Employing the XGBoost machine learning algorithm and lagged feature engineering to reflect real-world forecasting constraints, the analysis achieved strong predictive performance, with out-of-sample R² values of 0.8161 for Portugal and 0.7803 for Spain and corresponding Mean Absolute Errors (MAE) of 0.1703 and 0.1788, respectively. Short-term EURIBOR tenors (1-week, 1-month, 3-month) and the 12- month rate were the most influential predictors across both countries, showing their role in capturing market expectations about future monetary policy. Despite the similarities, important national differences were identified. In Spain, macroeconomic indicators such as GDP and unemployment exhibited a higher relative importance, suggesting that Spanish interest rate dynamics are more sensitive to broad economic conditions. In contrast, Portugal’s model revealed a stronger influence of demographic factors, indicating that structural population trends significantly shaped its interest rate environment. These findings carry theoretical and practical implications. The results affirm the differentiated transmission of common monetary policy across national markets within the Eurozone, and the study points out the need for localized forecasting models, particularly given the variations in mortgage structures. Portugal’s high prevalence of variable-rate mortgages contrasts Spain’s growing share of fixed-rate lending. The study demonstrates the power of combining machine learning techniques with economic theory to forecast key financial indicators, providing valuable insights for policymakers, investors, and financial institutions navigating an increasingly complex and heterogeneous European interest rate situation. | pt_PT |
| dc.identifier.tid | 203969014 | |
| dc.identifier.uri | http://hdl.handle.net/10362/184431 | |
| dc.language.iso | eng | pt_PT |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | pt_PT |
| dc.subject | EURIBOR | pt_PT |
| dc.subject | Housing Market | pt_PT |
| dc.subject | Cross Country Analysis | pt_PT |
| dc.subject | XGBoost | pt_PT |
| dc.subject | Optuna | pt_PT |
| dc.subject | SHAP Interpretability | pt_PT |
| dc.subject | SDG 8 - Decent work and economic growth | pt_PT |
| dc.subject | SDG 9 - Industry, innovation and infrastructure | pt_PT |
| dc.subject | SDG 10 - Reduced inequalities | pt_PT |
| dc.subject | SDG 11 - Sustainable cities and communities | pt_PT |
| dc.title | EURIBOR Fluctuations and Property Values: A Machine Learning Comparison of Portugal and Spain | pt_PT |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| rcaap.embargofct | The thesis was submitted as a paper to a top journal. | pt_PT |
| rcaap.rights | embargoedAccess | pt_PT |
| rcaap.type | masterThesis | pt_PT |
| thesis.degree.name | Mestrado em Marketing Analítico, especialização em Ciência de Dados Aplicada ao Marketing | pt_PT |