Can Twitter Forecast Uncertainty of Stocks?


Creative Commons License

Bozma G., Kul S.

SOSYOEKONOMI, cilt.28, sa.45, ss.315-326, 2020 (ESCI) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 28 Sayı: 45
  • Basım Tarihi: 2020
  • Doi Numarası: 10.17233/sosyoekonomi.2020.03.18
  • Dergi Adı: SOSYOEKONOMI
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.315-326
  • Atatürk Üniversitesi Adresli: Evet

Özet

Academic studies have shown that there is a relationship between emotional analysis results of tweets and stock price movements, and then stock prices can be estimated using this relationship. In this study, in which the effect of tweets on the volatility of the stock is estimated, the volatility scores and the emotion scores between the stocks were also revealed. In the scope of the study, sentiment analysis with Naive Bayes was performed on Turkish tweets shared by three phone companies (Alcatel, Turkcell and Vestel) which are in Borsa Istanbul and whose products are sold in Turkey. According to the results of the analysis, it was found that sentiment scores obtained for Turkcell and Vestel significantly increased Alcatel's conditional variance statistically.