Pearson correlation and granger causality analysis of twitter sentiments and the daily changes in bist30 index returns

Abstract

In this research an extensive study to identify the relations between Turkish financial-related tweets and the daily changes in Bist30 index returns is presented. After the sentiment analysis of Turkish financial-related tweets, the polarity values of the tweets are determined and the results of Pearson correlation and Granger causality analyses between the polarity values and the changes in Bist30 index returns are examined for two different time periods. The first research period covers the dates between 07.05.2018 and 30.04.2019. The second research period includes the dates between 07.05.2018 and 31.07.2018; where some important events took place in Turkey. To the best of our knowledge, this study is the most comprehensive study that analyses the relations between Turkish financial-related tweets and the daily changes in stock market returns. The framework of the study is shown in Figure A.