Working Paper: CEPR ID: DP10534
Authors: Bastian von Beschwitz; Donald B. Keim; Massimo Massa
Abstract: We investigate whether providers of high frequency news analytics affect the stock market. As identification, we exploit a unique experiment based on differences in news event classifications between different product releases of a major provider of news analytics. We document a causal effect of news analytics on the market, irrespective of the informational content of the news. Coverage in news analytics speeds up the market reaction in terms of stock price response and trading volume, and increases illiquidity immediately after the article. Furthermore, we document that traders learn dynamically about the precision of news analytics.
Keywords: information; institutional trading; stock price reaction; textual analysis
JEL Codes: G10; G12; G14
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| News Analytics (Y10) | Speed of Market Reactions (G14) |
| HRH Articles (Y30) | Speed of Stock Price Response (G17) |
| HRH Articles (Y30) | Increased Illiquidity (G19) |
| Ravenpack Sentiment Scores (G12) | Stock Returns (G12) |
| HRL Articles (J81) | Market Reactions (G10) |