mlsanityData to 7 Oct 2026

Realised Volatility Forecasting: Machine Learning via Financial Word Embedding

arXiv 2108.00480 ↗First posted 1 Aug 2021q-fin.CP cs.CL cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

We examine whether financial news can improve realised volatility forecasting using a parsimonious NLP-based framework that incorporates specialised financial word embeddings alongside general-purpose alternatives. News-only forecasts contain useful predictive information but generally do not outperform strong volatility-history benchmarks. Crucially, combining stock-related news forecasts with a strong volatility-history benchmark lowers forecast losses for several specifications and increases realised utility, providing evidence consistent with forecast complementarity. Performance varies across news types, embedding representations, and volatility regimes. SHAP attributions associate forecast variation with economically interpretable firm-specific and macroeconomic news themes.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 19 of its 67 references that are in the map. First posted on arXiv 1 Aug 2021.

Builds on

19 of 67 references.

References · 19

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
  4. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,462
  5. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  6. 2016Enriching Word Vectors with Subword Informationcited by 1,578
  7. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  8. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,124
  9. 2016FastText.zip: Compressing text classification modelscited by 331
  10. 2017Advances in Pre-Training Distributed Word Representationscited by 207
  11. 2014SimLex-999: Evaluating Semantic Models with (Genuine) Similarity Estimationcited by 170
  12. 2015A Primer on Neural Network Models for Natural Language Processingcited by 75
7 more
  1. 2019Deep Reinforcement Learning for Tradingcited by 73
  2. 2018Universal features of price formation in financial markets: perspectives from Deep Learningcited by 59
  3. 2019Enhancing Time Series Momentum Strategies Using Deep Neural Networkscited by 31
  4. 2021Slow Momentum with Fast Reversion: A Trading Strategy Using Deep Learning and Changepoint Detectioncited by 13
  5. 2020SHAP values for Explaining CNN-based Text Classification Modelscited by 12
  6. 2021Multi-Horizon Forecasting for Limit Order Books: Novel Deep Learning Approaches and Hardware Acceleration using Intelligent Processing Unitscited by 10
  7. 2020Building Cross-Sectional Systematic Strategies By Learning to Rankcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 1; 2024: 2; 2025: 1; 2026: 1.

7 papers.

  1. 2024Can LLMs Learn Macroeconomic Narratives from Social Media?cited by 7
  2. 2022DeepVol: Volatility Forecasting from High-Frequency Data with Dilated Causal Convolutionscited by 6
  3. 2022Word Embedding for Social Sciences: An Interdisciplinary Surveycited by 3
  4. 2025Fusing Narrative Semantics for Financial Volatility Forecastingcited by 3
  5. 2026Regime-aware financial volatility forecasting via in-context learningcited by 1
  6. 2023Deep Learning Enhanced Realized GARCHcited by 0
  7. 2024Optimal Text-Based Time-Series Indicescited by 0