Social-Media Sentiment, Portfolio Complexity, and Stock Returns

Abstract

Using tweets from StockTwits and machine-learning classification techniques, we find that social-media sentiment predicts positively and significantly future stock returns, and, importantly, such positive predictability decreases when the number of stocks users follow increases. The return predictability appears to stem from users’ ability to forecast future earnings. Further tests reveal that reduced predictability due to stock coverage is significant only for firms that are complex, opaque, and thus hard to analyze. The evidence suggests that stock analysis by users with a more complex portfolio is inferior due to attention and time constraints.

Keywords: Social-media sentiment, StockTwits, Behavioral finance, Limited attention, Stock returns

JEL Classification: G11, G12, G41