Fusing Narrative Semantics for Financial Volatility Forecasting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911228447686656 |
|---|---|
| author | Kong, Yaxuan Hwang, Yoontae Kaiser, Marcus Vryonides, Chris Oomen, Roel Zohren, Stefan |
| author_facet | Kong, Yaxuan Hwang, Yoontae Kaiser, Marcus Vryonides, Chris Oomen, Roel Zohren, Stefan |
| contents | We introduce M2VN: Multi-Modal Volatility Network, a novel deep learning-based framework for financial volatility forecasting that unifies time series features with unstructured news data. M2VN leverages the representational power of deep neural networks to address two key challenges in this domain: (i) aligning and fusing heterogeneous data modalities, numerical financial data and textual information, and (ii) mitigating look-ahead bias that can undermine the validity of financial models. To achieve this, M2VN combines open-source market features with news embeddings generated by Time Machine GPT, a recently introduced point-in-time LLM, ensuring temporal integrity. An auxiliary alignment loss is introduced to enhance the integration of structured and unstructured data within the deep learning architecture. Extensive experiments demonstrate that M2VN consistently outperforms existing baselines, underscoring its practical value for risk management and financial decision-making in dynamic markets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20699 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fusing Narrative Semantics for Financial Volatility Forecasting Kong, Yaxuan Hwang, Yoontae Kaiser, Marcus Vryonides, Chris Oomen, Roel Zohren, Stefan Computational Finance Artificial Intelligence We introduce M2VN: Multi-Modal Volatility Network, a novel deep learning-based framework for financial volatility forecasting that unifies time series features with unstructured news data. M2VN leverages the representational power of deep neural networks to address two key challenges in this domain: (i) aligning and fusing heterogeneous data modalities, numerical financial data and textual information, and (ii) mitigating look-ahead bias that can undermine the validity of financial models. To achieve this, M2VN combines open-source market features with news embeddings generated by Time Machine GPT, a recently introduced point-in-time LLM, ensuring temporal integrity. An auxiliary alignment loss is introduced to enhance the integration of structured and unstructured data within the deep learning architecture. Extensive experiments demonstrate that M2VN consistently outperforms existing baselines, underscoring its practical value for risk management and financial decision-making in dynamic markets. |
| title | Fusing Narrative Semantics for Financial Volatility Forecasting |
| topic | Computational Finance Artificial Intelligence |
| url | https://arxiv.org/abs/2510.20699 |