An Explorative Analysis of Tweets Sentiments for Investment Decision in Stock Markets

Abstract

Nowadays lots of researches report that positive or negative social media posts have significant effects on stock market returns. Unfortunately, the number of studies conducted in this context for Turkey is very limited. Therefore, this study investigates the relationship between Turkish tweets and the trend of Bist30 index returns. Correlation analysis is a statistical method used to measure the strength and direction of relationship between two variables. The existence of a correlation between sentiments of financial tweets and the stock market returns may be an indicator that social media can be used as a resource to predict the direction or values of stock markets returns. In this study, the correlation analysis steps are conducted on the polarity scores and changes in Bist30 index returns with all the necessary statistical tests. The contribution of our study is that analyzes are made on a daily, weekly and monthly basis. The experimental results show that there are significant positive correlations between the sentiment polarity values and changes in Bist30 index returns. As a result, this study points out a useful pathway for the future researches to show that social media posts may convey useful information for financial markets.

Keywords

• Correlation analysis
• sentiment analysis
• Turkish tweets
• Bist30

References

1. [1] Liu, L. X., Ann E. S., Yong Z., The long-run role of the media: Evidence from initial public offerings, Management Science, 60 (8), 1945-1964, 2014.
2. [2] Atan, S., Çinar, Y. Borsa Istanbul’da finansal haberler ile piyasa degeri iliskisinin metin madenciligi ve duygu (sentiment) analizi ile incelenmesi, Ankara Üniversitesi Sbf Dergisi,74 (1),1-34, 2019.
3. [3] Teti, E., Dallocchio, M., Aniasi, A. The relationship between twitter and stock prices. Evidence from the US technology industry, Technological Forecasting and Social Change, 149 (119747), 1-9, 2019.
4. [4] Sprenger T. O., Tumasjan A., Sandner P. G., Welpe, I. M., Tweets and trades: The information content of stock microblogs, European Financial Management, 20 (5), 926-957, 2014.
5. [5] Tabari, N., Biswas, P., Praneeth, B., Seyeditabari, A., Hadzikadic, M., Zadrozny, W., Causality analysis of twitter sentiments and stock market returns. In Proceedings of the First Workshop on Economics and Natural Language Processing, 11-19, 2018.
6. [6] Eliaçik A. B., Erdogan N., Mikro Bloglardaki Finans Topluluklari için Kullanici Agirliklandirilmis Duygu Analizi Yöntemi, Ulusal Yazılım Mühendisliği Sempozyumu, İzmir, 782-793, 2015.
7. [7] Yildirim M., Yüksel C. A., Sosyal Medya ile Hisse Senedi Fiyatinin Günlük Hareket Yönü Arasindaki Iliskinin Incelenmesi: Duygu Analizi Uygulamasi. Uluslararasi Iktisadi ve Idari Incelemeler Dergisi, 22, 33-44, 2017.
8. [8] Ates, E., Guran, A. (2021). Pearson correlation and Granger causality analysis of Twitter sentiments and the daily changes in Bist30 index returns. Journal of the Faculty of Engineering and Architecture of Gazi University, 36(3), 1688-1701.

9. [9] Bollen J., Mao H., Zeng X., Twitter mood predicts the stock market, Journal of computational science, 2 (1), 1-8, 2011.
10. [10] Mittal A., Goel A., Stock prediction using twitter sentiment analysis, Working Paper Standford University, CS 229, 1-5, 2012.
11. [11] Ranco G., Aleksovski D., Caldarelli G., Grcar M., Mozetic I., The effects of Twitter sentiment on stock price returns, PloS one, 10 (9), 1-21, 2015.
12. [12] Pagolu V. S., Reddy K. N., Panda G., Majhi B, Sentiment analysis of Twitter data for predicting stock market movements, In 2016 international conference on signal processing, communication, power and embedded system (SCOPES), Paralakhemundi-Hindistan, 1345-1350, 3-5 Ekim, 2016.
13. [13] Deng S., Huang Z. J., Sinha A. P., Zhao H., The Interaction between Microblog Sentiment and Stock Return: An Empirical Examination, MIS quarterly, 42 (3), 895-918, 2018.
14. [14] Zhao R., Quantifying the correlation and prediction of daily happiness sentiment and stock return: The Case of Singapore, Physica A: Statistical Mechanics and its Applications, 533, 1-9, 2019.
15. [15] https://libguides.library.kent.edu/spss/pearsoncorr
16. [16] Bland M. 4th ed. Oxford: Oxford University Press; 2015. An Introduction to Medical Statistics. [Google Scholar] [Ref list]
17. [17] Campbell MJ, Machin D, Walters SJ. 4th ed. Chichester: John Wiley & Sons, Ltd; 2007. Medical Statistics: A text book for the health sciences. [Google Scholar] [Ref list]
18. [18] Mishra, P., Pandey, C. M., Singh, U., Gupta, A., Sahu, C., & Keshri, A. (2019). Descriptive statistics and normality tests for statistical data. Annals of cardiac anaesthesia, 22(1), 67.
19. [19] Ghasemi A, Zahediasl S. Normality tests for statistical analysis: A guide for non-statisticians. Int J Endocrinol Metab. 2012;10:486–9. [PMC free article] [PubMed] [Google Scholar]
20. [20] Kim HY. Statistical notes for clinical researchers: Assessing normal distribution (2) using skewness and kurtosis. Restor Dent Endod. 2013;38:52–4. [PMC free article] [PubMed] [Google Scholar]
21. [21] 12. Armitage P, Berry G. 2nd ed. London: Blackwell Scientific Publications; 1987. Statistical Methods in Medical Research. [Google Scholar]
22. [22] 13. Barton B, Peat J. 2nd ed. Sydney: Wiley Blackwell, BMJ Books; 2014. Medical Statistics: A Guide to SPSS, Data Analysis and Clinical Appraisal. [Google Scholar]
23. [23] IBM SPPS yazılımı. https://www.ibm.com/tr-tr/analytics/spss-statistics-software. Erişim tarihi: Haziran, 1, 2019.