Analyzing the influence of twitter hashtags on the movement of the stock market

Abstract

This paper evaluates the existence of a relationship between a well-known micro-blogging platform, Twitter and the financial market. We particularly consider in a period of six months the daily collection of trending twitter hash tags and the daily movement of the NSE 20 share index Granger causality, Correlation tests and logistic regression tests were used to examine the relationship. We find low Pearson correlation between the corresponding time series over the time period. We also found that none of the variables granger causes the other. We found no relationship between the sentiment as expressed from twitter and the movement of the stock market. As a series of other papers have already shown, there is a signal worth investigating which connects social media and market behavior. This opens the way, if not to forecasting, then at least to "now-casting" financial markets. This study is open to further and more advanced research using a longer time period as well and to employ sentiment classification techniques.