Quantifying high-frequency market reactions to real-time news sentiment announcements

Arbeitspapier

We examine intra-day market reactions to news in stock-specific sentiment disclosures. Using pre-processed data from an automated news analytics tool based on linguistic pattern recognition we extract information on the relevance as well as the direction of company-specific news. Information-implied reactions in returns, volatility as well as liquidity demand and supply are quantified by a high-frequency VAR model using 20 second intervals. Analyzing a cross-section of stocks traded at the London Stock Exchange (LSE), we find market-wide robust news-dependent responses in volatility and trading volume. However, this is only true if news items are classified as highly relevant. Liquidity supply reacts less distinctly due to a stronger influence of idiosyncratic noise. Furthermore, evidence for abnormal highfrequency returns after news in sentiments is shown.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: SFB 649 Discussion Paper ; No. 2009,063

Classification and Topics

• Classification: Wirtschaft ; Information and Market Efficiency; Event Studies; Insider Trading ; Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes; State Space Models
• Subject: firm-specific news ; news sentiment ; high-frequency data ; volatility ; liquidity ; abnormal returns ; Börsenkurs ; Kapitalertrag ; Volatilität ; Ankündigungseffekt ; Publizitätspflicht ; Informationseffizienz ; Marktliquidität ; Schätzung ; Großbritannien

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Groß-Klußmann, Axel ; Hautsch, Nikolaus
• Event: Veröffentlichung
• (who): Humboldt University of Berlin, Collaborative Research Center 649 - Economic Risk
• (where): Berlin
• (when): 2009

Further information

• Handle: http://hdl.handle.net/10419/39334
• Last update: 10.03.2025, 11:43 AM CET