How do big markets react to investors’ sentiments on firm tweets?

ABSTRACT

Social media actors can disseminate their views about a firm, creating potential consequences for the firm. This study thus aims to explore the reaction of a firm stock market in response to the sentiments of investors on a firm's financial tweets. It focuses on the investors’ sentiments of the world's largest 100 firms’ tweets. The Naïve Bayesian classification method is used to classify sentiments of tweets into their positive and negative tones. The cumulative abnormal return and the buy-and-hold return are used as proxies for the market reactions. The study finds that positive (negative) Twitter sentiments are likely to affect positively (negatively) the market cumulative abnormal return and its buy-and-hold return. This effect is likely to hold for a shorter period and diminishes over a longer time. Besides, this effect is more associated with negative sentiments than positive ones. Further, the analysis indicates that the effect of positive (negative) sentiments is more observable in common (coded) law countries. The results suggest that Twitter sentiments carry information contents indicative of the stock market, asserting the substantial role of information on social media. The results call for establishing a direct link between social network platforms and stock markets to mitigate the panic diffusion in the market that may arise from sentiments of investors.

KEYWORDS:

• Social media disclosure
• financial information
• Twitter sentiments
• stock markets
• cumulative abnormal return
• buy-and-hold return
• largest 100 firms

Acknowledgments

We would like to thank Prof. Olaf Weber (the editor of the Journal of Sustainable Finance & Investment) and the anonymous referees for their helpful comments and suggestions, which substantially improved the quality of the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For details, please check the following: https://www.cs.stanford.edu/people/alecmgo/papers/TwitterDistantSupervision09.pdf