Trading on Twitter: Using Social Media Sentiment to Predict Stock Returns

摘要: Decision making is often based on the rational assessment of information, but recent research shows that emotional sentiment also plays an important role, especially for investment decision making. Emotional about a firm's stock spreads rapidly through social media more likely to be incorporated quickly into prices (e.g., same trading day it was expressed), while slowly takes longer and thus predict future days. We analyzed cumulative (positive negative) in 2.5 million Twitter postings individual S&P 500 firms compared this returns those firms. Our results show tweets specific firm from users with less than 171 followers (the median our sample) had significant impact stock's next day, 10 days, 20 Interestingly, fewer were not retweeted greatest returns. A strategy these findings produced meaningful economic gains order 11–15% annual return.