Paper: arXiv 2410.10614

Abstract

The diffusion of financial news into market prices is a complex process, making it challenging to evaluate the connections between news events and market movements. This paper introduces FININ (Financial Interconnected News Influence Network), a novel market prediction model that captures not only the links between news and prices but also the interactions among news items themselves. FININ effectively integrates multi-modal information from both market data and news articles. We conduct extensive experiments on two datasets, encompassing the S&P 500 and NASDAQ 100 indices over a 15-year period and over 2.7 million news articles. The results demonstrate FININ’s effectiveness, outperforming advanced market prediction models with an improvement of 0.429 and 0.341 in the daily Sharpe ratio for the two markets respectively. Moreover, our results reveal insights into the financial news, including the delayed market pricing of news, the long memory effect of news, and the limitations of financial sentiment analysis in fully extracting predictive power from news data.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 6.0/10 (capped from 8.0: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a novel neural network architecture (FININ) with multi-modal fusion and attention mechanisms, involving moderate-to-advanced mathematical modeling, but grounds its claims in extensive empirical backtesting on two major indices over 15 years with 2.7 million news articles and reports significant Sharpe ratio improvements.

Research Flowchart

  flowchart TD
  A["Research Goal: Modeling News-Price Diffusion for Market Prediction"] --> B["Data Acquisition & Processing"]
  B --> C["FININ Architecture: Multi-modal Integration"]
  C --> D["Computational Process: News Influence Network"]
  D --> E["Model Training & Validation"]
  E --> F["Key Outcomes: Enhanced Sharpe Ratio & Market Insights"]
  
  subgraph Data Inputs
      B --> B1["2.7M News Articles"]
      B --> B2["S&P 500 & NASDAQ Data"]
  end
  
  subgraph Computational Layers
      D --> D1["News-News Interactions"]
      D --> D2["News-Price Links"]
      D --> D3["Temporal Diffusion Modeling"]
  end
  
  subgraph Findings
      F --> F1["+0.429 Sharpe Ratio (S&P 500)"]
      F --> F2["+0.341 Sharpe Ratio (NASDAQ 100)"]
      F --> F3["Delayed Market Pricing"]
      F --> F4["Long Memory Effect of News"]
  end

Cited References (107)

  1. How the equity market responds to unanticipated events (2003)
  2. Do markets punish left governments? (2013)
  3. Measuring the effects of federal reserve forward guidance and asset purchases on financial markets (2021)
  4. Relative importance of scheduled macroeconomic news for stock market investors (2003)
  5. Popular news are relevant news! How investor attention affects algorithmic decision-making and decision support in financial markets (2021)
  6. Macro news and micro news: complements or substitutes? (2022)
  7. Financial news recommendation based on graph embeddings (2019)
  8. Understanding Financial Information Seeking Behavior from User Interactions with Company Filings (2022)
and 99 more
  1. Investor sentiment aligned: A powerful predictor of stock returns (2015)
  2. Joint news, attention spillover, and market returns (2021)
  3. Stock price reaction to news: The joint effect of tone and attention on momentum (2017)
  4. Hybrid learning to rank for financial event ranking (2021)
  5. From ranknet to lambdarank to lambdamart: An overview (2010)
  6. Lexical Event Ordering with an Edge-Factored Model (2015)
  7. Irgan: A minimax game for unifying generative and discriminative information retrieval models (2017)
  8. Deep learning with long short-term memory networks for financial market predictions (2018)
  9. Deep learning stock volatility with google domestic trends (2015) — link
  10. Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500 (2017)
  11. Applications of news analytics in finance: A review (2011)
  12. FinBERT : Financial sentiment analysis with pre-trained language models (2019) — link
  13. Giving content to investor sentiment: The role of media in the stock market (2007)
  14. How news and its context drive risk and returns around the world (2019)
  15. The exchange rate effects of macro news after the global financial crisis (2019)
  16. FAST: Financial news and tweet based time aware network for stock trading (2021)
  17. Ranking financial tweets (2016)
  18. Learning to rank for information retrieval and natural language processing (2014)
  19. Learning to rank for information retrieval (2009)
  20. Albert: A lite BERT for self-supervised learning of language representations (2019) — link
  21. Learning to rank: from pairwise approach to listwise approach (2007)
  22. Learning to rank for recommender systems (2013)
  23. Learning to rank question answer pairs with holographic dual lstm architecture (2017)
  24. Mcrank: Learning to rank using multiple classification and gradient boosting (2007)
  25. Learning to rank using gradient descent (2005)
  26. Ranking measures and loss functions in learning to rank (2009)
  27. Personalized re-ranking for recommendation (2019)
  28. BloombergGPT: A Large Language Model for Finance (2023) — link
  29. Topic-dependent sentiment analysis of financial blogs (2009)
  30. Macroeconomics news and the stock market: Evidence from Europe (1989)
  31. Stock market overreactions to bad news in good times: a rational expectations equilibrium model (1999)
  32. Evaluation of sentiment analysis in finance: from lexicons to transformers (2020)
  33. Internet investment advice: Investing with a rock of salt (2003)
  34. When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks (2011)
  35. News impact on stock price return via sentiment analysis (2014)
  36. Combining the wisdom of crowds and technical analysis for financial market prediction using deep random subspace ensembles (2018)
  37. Deep learning for financial sentiment analysis on finance news providers (2016)
  38. Realization of a news dissemination agent based on weighted association rules and text mining techniques (2010)
  39. Multi-modal Attention Network for Stock Movements Prediction (2021) — link
  40. The information content of forward-looking statements in corporate filings—A na "\i ve Bayesian machine learning approach (2010)
  41. Forex-foreteller: Currency trend modeling using news articles (2013)
  42. Crowds on wall street: Extracting value from social investing platforms (2014) — link
  43. Fine-grained sentiment analysis on financial microblogs and news headlines (2017)
  44. BERT : Pre-training of deep bidirectional transformers for language understanding (2018) — link
  45. Attention is all you need (2017)
  46. A hybrid model integrating deep learning with investor sentiment analysis for stock price prediction (2021)
  47. Applications of deep learning in stock market prediction: recent progress (2021)
  48. An Empirical Survey on Long Document Summarization: Datasets, Models, and Metrics (2022)
  49. Market response to analyst recommendations in the “dartboard” column: the information and price-pressure effects (1996)
  50. Market responses to buy recommendations issued by Personal Finance Magazines: effects of information, price-pressure, and company characteristics (2007)
  51. How does news sentiment impact asset volatility? Evidence from long memory and regime-switching approaches (2013)
  52. Attention based multi-modal new product sales time-series forecasting (2020)
  53. A multimodal event-driven lstm model for stock prediction using online news (2020)
  54. All that glitters: The effect of attention and news on the buying behavior of individual and institutional investors (2008)
  55. Stock movement prediction from tweets and historical prices (2018)
  56. The role of text-extracted investor sentiment in Chinese stock price prediction with the enhancement of deep learning (2020)
  57. A novel ensemble deep learning model for stock prediction based on stock prices and news (2022)
  58. A hierarchical attention network for stock prediction based on attentive multi-view news learning (2022)
  59. Information diffusion, cluster formation and entropy-based network dynamics in equity and commodity markets (2017)
  60. Deep neural networks for youtube recommendations (2016)
  61. Rankt5: Fine-tuning T5 for text ranking with ranking losses (2023)
  62. A deep relevance matching model for ad-hoc retrieval (2016)
  63. Trading and information diffusion in over-the-counter markets (2018)
  64. Stock market response to information diffusion through internet sources: A literature review (2019)
  65. Online big data-driven oil consumption forecasting with Google trends (2019)
  66. Stock market prices and long-range dependence (1999)
  67. Long-range dependence in daily stock volatilities (2000)
  68. The effect of long memory in volatility on stock market fluctuations (2007)
  69. RoBERTa : A robustly optimized BERT pretraining approach (2019) — link
  70. Finbert: A pre-trained financial language representation model for financial text mining (2021)
  71. C-pack: Packaged resources to advance general chinese embedding (2023) — link
  72. MTEB: Massive text embedding benchmark (2022) — link
  73. Llama 2: Open foundation and fine-tuned chat models (2023) — link
  74. One Fits All: Power General Time Series Analysis by Pretrained LM (2023) — link
  75. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models (2023)
  76. Reinforcement-learning based portfolio management with augmented asset movement prediction states (2020)
  77. Predicting Stock Trends with CNN-BiLSTM Based Multi-Feature Integration Model (2021)
  78. Transformation of dense and sparse text representations (2019) — link
  79. Scaling sentence embeddings with large language models (2023) — link
  80. Analysis of relationships between tweets and stock market trends (2018)
  81. News sensitive stock trend prediction (2002)
  82. How news affects the trading behaviour of different categories of investors in a financial market (2015)
  83. Causality-Guided Multi-Memory Interaction Network for Multivariate Stock Price Movement Prediction (2023)
  84. Cmg: A class-mixed generation approach to out-of-distribution detection (2022)
  85. Stock price prediction using news sentiment analysis (2019)
  86. Hrn: A holistic approach to one class learning (2020)
  87. Stock trend prediction with multi-granularity data: A contrastive learning approach with adaptive fusion (2021)
  88. Temporal and heterogeneous graph neural network for financial time series prediction — → 2305.08740
  89. Price forecast with high-frequency finance data: An autoregressive recurrent neural network model with technical indicators (2020)
  90. Hierarchical Multi-Scale Gaussian Transformer for Stock Movement Prediction. (2020)
  91. Efficient capital markets: A review of theory and empirical work (1970)
  92. Market reaction to internet news: Information diffusion and price pressure (2016)
  93. PEN: prediction-explanation network to forecast stock price movement with better explainability (2023)
  94. Multilingual Clustering of Streaming News (2018)
  95. Gpt-4 technical report (2023) — link
  96. MANA-Net: Mitigating Aggregated Sentiment Homogenization with News Weighting for Enhanced Market Prediction — → 2409.05698
  97. Belt: A pipeline for stock price prediction using news (2020)
  98. Comparing the effectiveness of deep feedforward neural networks and shallow architectures for predicting stock price indices (2020)
  99. Rest: Relational event-driven stock trend forecasting (2021)

2 of 107 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.