A Large Scale Study to Understand the Relation between Twitter and Financial Market

摘要

Twitter has transformed from an online platform for communication to a mega content generator for all kinds of topics. The topic of posts (or tweets) generated on Twitter cover diverse topics of interests. For example, politics, public personalities, events and corporate organizations. In this study, we analysed if tweets related to corporate organizations can predict the financial market. Our analysis is performed on a Twitter dataset which spans over more than two years and is related to more than 1723 stocks. To the best of our knowledge this amount of large and big dataset, specifically in terms of time length and stocks has not been studied in the past. Our quantitative analysis shows that on an average correlation between tweets and stocks' volume being traded is 0.29 on an average. In this empirical study, we also evaluated the stocks from Yahoo's sectors' categories perspective to find which sectors are more correlated than others. We also looked at the influence of important users, that is users with a large number of followers. Our results ally with the fact that important users contribute more in influencing the market [3] rather than the wisdom of the crowd [6]. The verification of our results using statistical approaches on a large dataset can be seen as a contribution in the area of financial studies using data from online platforms.