Media Content and Stock Returns: The Predictive Power of Press

• DOI: 10.17578/19-1-1
• Publication Date: 2015-06-12T03:45:57Z
• Authors: Nicky J. Ferguson; ; Dennis Philip; ; Herbert Y. T. Lam; ; Jie Michael Guo
• Keywords: Stock (firearms); Predictive power; News analytics; Tone (literature); Media Coverage

This paper examines whether tone (positive and negative) volume of firm-specific news media content provide valuable information about future stock returns, using UK data from 1981–2010. The results indicate that both significantly predict next period abnormal with the impact more pronounced than tone. Additionally, predictive power is found to be stronger among lower visibility firms. Further, finds evidence an attention-grabbing effect for stories high coverage, mainly seen larger A simple news-based trading strategy produces statistically significant risk-adjusted returns 14.2 19 basis points in 2003–2010. At aggregate level, price pressure induced by semantics corrected only part subsequent reversals. Overall, findings suggest incorporates predicts asset returns.