Information or noise: How Twitter facilitates stock market information aggregation

Florian Kiesel, Thomas Pöppe, Sascha Kolaric, Dirk Schiereck

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

We assess the relevance of Twitter for stock-relevant information dissemination in financial markets on the single stock level. We use a unique dataset including more than 12 million Twitter feeds linked to specific firms. Using intraday data for the computation of advanced trading metrics, such as effective spreads, intraday volatility, and a daily version of the microstructure variable probability of informed trading (PIN), we measure the impact of Twitter activity on trading and information dissemination. The PIN model indicates that more uninformed than informed traders rush to the market along with rising Twitter activity. These results indicate that Twitter serves as an excellent indicator of news that is relevant for the stock market. However, we show that Twitter does not lead traditional news channels. In contrast, Twitter activity follows the market and has no predictive power with regard to future stock trading volume or volatility on the single stock level.
Original languageEnglish
Title of host publicationICIS 2019 Proceedings
Publication statusPublished - 31 Dec 2019

Keywords / Materials (for Non-textual outputs)

  • social media
  • Twitter
  • microblogging
  • probability of informed trading (PIN)
  • iInformation dissemination
  • information asymmetry

Fingerprint

Dive into the research topics of 'Information or noise: How Twitter facilitates stock market information aggregation'. Together they form a unique fingerprint.

Cite this