Journals / Erciyes Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi / 2026 / Sayı: 74

Stock market reaction prediction using social media-based sentiment analysis: An application using X data

Sosyal medya tabanlı duygu analizi ile borsa tepki tahmini: X verileri üzerine bir uygulama

Pages
9–17
Publication Date
2026-08-31
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Regression results suggest that average sentiment is positively and statistically significantly related to daily returns, although the economic magnitude is modest.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study examines whether social-media-based investor sentiment is associated with stock market performance using data from X (formerly Twitter).
📊
Key Findings & Data:Regression results suggest that average sentiment is positively and statistically significantly related to daily returns, although the economic magnitude is modest.
💡
Conclusion & Impact:These results imply that social-media-derived sentiment is best viewed as a complementary input to traditional analysis rather than a standalone trading rule, and that monitoring increases in negative sentiment may be informative for short-term risk assessment.
🔬Design:Araştırma Makalesi
📍Scope:Comprehensive Sample
⚙️Variables:duygu analizi · hisse senedi getirileri · sosyal medya

Abstract

This study examines whether social-media-based investor sentiment is associated with stock market performance using data from X (formerly Twitter). Specifically, we test whether sentiment indicators derived from stock-specific posts are related to daily stock returns and abnormal returns in a panel setting. To this end, we compute stock day measures of (i) average sentiment, (ii) the shares of positive and negative messages, and (iii) tweet volume, and merge them with daily price data for a sample of stocks. We estimate fixed-effects panel regressions including stock and day effects, and report stock-clustered robust standard errors. Descriptive statistics indicate that posts about the sampled stocks are, on average, positively toned, with the share of positive messages exceeding that of negative ones. Regression results suggest that average sentiment is positively and statistically significantly related to daily returns, although the economic magnitude is modest. For abnormal returns, the share of negative sentiment is negative and statistically significant, while the positive sentiment share and tweet volume show weaker and less robust associations. Overall, the findings indicate that X-based sentiment measures provide a weak yet systematic signal that co-moves with stock performance, with negative sentiment playing a more prominent role in abnormal returns than positive sentiment. These results imply that social-media-derived sentiment is best viewed as a complementary input to traditional analysis rather than a standalone trading rule, and that monitoring increases in negative sentiment may be informative for short-term risk assessment.

Özet

Bu çalışma, sosyal medya temelli yatırımcı duyarlılığı göstergelerinin hisse senedi performansı ile ilişkisini X (eski adıyla Twitter) verileri üzerinden incelemektedir. Çalışmanın amacı, hisse bazlı X paylaşımlarından türetilen duygu göstergelerinin günlük hisse senedi getirileri ve anormal getirilerle ilişkisini panel veri çerçevesinde test etmektir. Bu kapsamda, seçili hisse senetlerine ilişkin X paylaşımlarından günlük ortalama duygu skoru, pozitif ve negatif duygu oranları ile tweet hacmi hesaplanmış; söz konusu göstergeler hisse–gün düzeyindeki fiyat verileriyle eşleştirilmiştir. Analizlerde hisse sabit etkileri ve gün sabit etkilerini içeren panel regresyon modelleri tahmin edilmiş; standart hatalar hisse düzeyinde kümelemeye (cluster-robust) göre raporlanmıştır. Betimsel bulgular, örneklemde ilgili hisselere yönelik paylaşımların genel olarak pozitif tonda seyrettiğini ve pozitif duygu oranının negatif duygu oranından yüksek olduğunu göstermektedir. Regresyon sonuçları, ortalama duygu skorunun günlük getirilerle pozitif yönde ve istatistiksel olarak anlamlı biçimde ilişkili olduğunu; ancak ilişkinin ekonomik büyüklüğünün sınırlı kaldığını ortaya koymaktadır. Anormal getiriler açısından negatif duygu oranı anlamlı ve negatif işaretli bulunurken, pozitif duygu oranı ve tweet hacmi daha zayıf sonuçlar vermektedir. Genel olarak bulgular, X tabanlı duygu göstergelerinin fiyat davranışına zayıf fakat sistematik bir sinyal düzeyinde eşlik edebildiğini ve olumsuz duyarlılığın anormal getirilerle ilişkisinin daha belirgin olduğunu göstermektedir. Sonuçlar, sosyal medya duyarlılığının geleneksel analizlere tamamlayıcı bir girdi olarak değerlendirilebileceğine işaret etmektedir.

🔖

Cite This Article

İmamoğlu, Ö., *, Ö. İ. (2026). Stock market reaction prediction using social media-based sentiment analysis: An application using X data. Erciyes Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, Sayı: 74, 9–17. https://doi.org/10.18070/erciyesiibd.1865123

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (30)

  1. Ackert, L. F., Church, B. K., & Deaves, R. (2003). Emotion and financial markets. Economic Review, 88(2), 33–41.
  2. Al Jazeera. (2022, October 28). How Elon Musk financed his $44bn Twitter takeover. https:// www.aljazeera.com/economy/2022/10/28/how-elon-musk-financed-his-twitter takeover
  3. Aydın, Ü., & Ağan, B. (2016). Rasyonel olmayan kararların finansal yatırım tercihleri üzerindeki etkisi: Davranışsal finans çerçevesinde bir uygulama. Ekonomik ve Sosyal Araştırmalar Dergisi, 12(2), 95–112.
  4. Baker, M., & Wurgler, J. (2006). Investor sentiment and the cross-section of stock returns. The Journal of Finance, 61(4), 1645–1680. https://doi.org/10.1111/j.1540 6261.2006.00885.x
  5. Bulunmaz, B. (2011). Social media usage in the automobile sector and the Fiat case. Global Media Journal: Turkish Edition, 2(3), 29–43.
  6. Cam, H., Cam, A. V., Demirel, U., & Ahmed, S. (2024). Sentiment analysis of financial Twitter posts with machine learning classifiers. Heliyon, 10(1), e23784. https://doi. org/10.1016/j.heliyon.2023.e23784
  7. Costa, D. F., Carvalho, F. de M., & Moreira, B. C. de M. (2019). Behavioral economics and behavioral finance: A bibliometric analysis of the scientific fields. Journal of Economic Surveys, 33(1), 3–24. https://doi.org/10.1111/joes.12262
  8. Dabner, N. (2012). “Breaking ground” in the use of social media: A case study of a university earthquake response to inform educational design with Facebook. The Internet and Higher Education, 15(1), 69–78. https://doi.org/10.1016/j.iheduc.2011.06.001
  9. Dawson, C., & Henley, A. (2012). “Push” versus “pull” entrepreneurship: An ambiguous distinction? International Journal of Entrepreneurial Behavior & Research, 18(6), 697–719.
  10. De Bondt, W. F. M., & Thaler, R. (1985). Does the stock market overreact? The Journal of Finance, 40(3), 793–805. https://doi.org/10.1111/j.1540-6261.1985.tb05004.x
  11. De Bondt, W. F. M., & Thaler, R. H. (1987). Further evidence on investor overreaction and stock market seasonality. The Journal of Finance, 42(3), 557–581. https://doi. org/10.1111/j.1540-6261.1987.tb04569.x
  12. Dowling, M., & Lucey, B. (2010). Other behavioral biases. İçinde H. K. Baker & J. R. Nofsinger (Ed), Behavioral finance: Investors, corporations, and markets (ss. 313 330). John Wiley & Sons.
  13. Equinxx. (2022). Stock tweets for sentiment analysis and prediction [Data set]. Kaggle.
  14. MacArthur, A. (2024, November 25). From Dorsey’s first tweet to Musk’s ownership: A comprehensive history of Twitter. Lifewire.
  15. Malmendier, U., & Tate, G. (2005). CEO overconfidence and corporate investment. The Journal of Finance, 60(6), 2661–2700. https://doi.org/10.1111/j.1540 6261.2005.00813.x
  16. Medetoğlu, B., & Saldanlı, A. (2019). Sürü davranışının hisse senedi fiyat hareketliliğine etkisi: BIST 100 örneği. İşletme Araştırmaları Dergisi, 11(2), 1191–1204.
  17. Mehta, I. (2022, October 28). Twitter will be delisted from the New York Stock Exchange on November 8. TechCrunch.
  18. Michailova, J. (2010). Development of the overconfidence measurement instrument for the economic experiment (MPRA Paper No. 4). Munich Personal RePEc Archive.
  19. Miller, M. H. (2000). The history of finance: An eyewitness account. Journal of Applied Corporate Finance, 12(2), 8–14. https://doi.org/10.1111/j.1745-6622.2000.tb00050.x
  20. Özcan, M. (2021). Sosyal medya platformlarının hisse senedi piyasalarına etkisi: BIST30 örneği. Gümüşhane Üniversitesi Sosyal Bilimler Enstitüsü Elektronik Dergisi, 12(1), 160–172.
  21. Plume, C. J., Reid, L. F., & Parker, R. S. (2017). Social media in the marketing context. Chandos Publishing.
  22. Pompian, M. M. (2006). Behavioral finance and wealth management: How to build optimal portfolios that account for investor biases. John Wiley & Sons.
  23. Ranco, G., Aleksovski, D., Caldarelli, G., Grčar, M., & Mozetič, I. (2015). The effects of Twitter sentiment on stock price returns. PLOS ONE, 10(9), e0138441. https://doi. org/10.1371/journal.pone.0138441
  24. Ritter, J. R. (2003). Behavioral finance. Pacific-Basin Finance Journal, 11(4), 429–437. https://doi.org/10.1016/S0927-538X(03)00048-9
  25. Rötheli, T. F. (2005). The illusion of over-optimism in survey data: The case of manufacturers’ selling prices. The Journal of Socio-Economics, 34(2), 151–159. https://doi.org/10.1016/j.socec.2004.09.005
  26. Rubinstein, M. (2002). Markowitz’s “portfolio selection”: A fifty-year retrospective. The Journal of Finance, 57(3), 1041–1045. https://doi.org/10.1111/1540-6261.00453
  27. Souza, T. T. P., Kolchyna, O., Treleaven, P. C., & Aste, T. (2015). Twitter sentiment analysis applied to finance: A case study in the retail industry. arXiv. http://arxiv.org/ abs/1507.00784
  28. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. https://doi.org/10.1126/science.185.4157.1124
  29. Yıldırım, M., & Yüksel, C. A. (2017). Sosyal medya ile hisse senedi fiyatının günlük hareket yönü arasındaki ilişkinin incelenmesi: Duygu analizi uygulaması. Uluslararası İktisadi ve İdari İncelemeler Dergisi, (22. UPK Ahmet Hamdi İslamoğlu Özel Sayısı), 33–44. https://doi.org/10.18092/ulikidince.352414
  30. Zheludev, I., Smith, R., & Aste, T. (2014). When can social media lead financial markets? Scientific Reports, 4, 4213. https://doi.org/10.1038/srep04213