Federated Learning for Financial Forecasting
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915504584654848 |
|---|---|
| author | Noseda, Manuel De Luca, Alberto Von Briel, Lukas Lacour, Nathan |
| author_facet | Noseda, Manuel De Luca, Alberto Von Briel, Lukas Lacour, Nathan |
| contents | This paper studies Federated Learning (FL) for binary classification of volatile financial market trends. Using a shared Long Short-Term Memory (LSTM) classifier, we compare three scenarios: (i) a centralized model trained on the union of all data, (ii) a single-agent model trained on an individual data subset, and (iii) a privacy-preserving FL collaboration in which agents exchange only model updates, never raw data. We then extend the study with additional market features, deliberately introducing not independent and identically distributed data (non-IID) across agents, personalized FL and employing differential privacy. Our numerical experiments show that FL achieves accuracy and generalization on par with the centralized baseline, while significantly outperforming the single-agent model. The results show that collaborative, privacy-preserving learning provides collective tangible value in finance, even under realistic data heterogeneity and personalization requirements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16393 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Federated Learning for Financial Forecasting Noseda, Manuel De Luca, Alberto Von Briel, Lukas Lacour, Nathan Machine Learning Applications This paper studies Federated Learning (FL) for binary classification of volatile financial market trends. Using a shared Long Short-Term Memory (LSTM) classifier, we compare three scenarios: (i) a centralized model trained on the union of all data, (ii) a single-agent model trained on an individual data subset, and (iii) a privacy-preserving FL collaboration in which agents exchange only model updates, never raw data. We then extend the study with additional market features, deliberately introducing not independent and identically distributed data (non-IID) across agents, personalized FL and employing differential privacy. Our numerical experiments show that FL achieves accuracy and generalization on par with the centralized baseline, while significantly outperforming the single-agent model. The results show that collaborative, privacy-preserving learning provides collective tangible value in finance, even under realistic data heterogeneity and personalization requirements. |
| title | Federated Learning for Financial Forecasting |
| topic | Machine Learning Applications |
| url | https://arxiv.org/abs/2509.16393 |