Social Media Sentiment Stock Market
- Published:
Social media sentiment and the stock market
Journal of Economic science and Finance volume 46,pages 397–419 (2022)Cite this commodity
Abstract
This paper investigates the link between social media sentiment and the stock marketplace using more than two one thousand thousand tweets posted in 2017 that include the name or symbol of 20-v companies listed in the S&P 100. We find a ii-way relationship when using hourly and daily intervals: We notice that a college proportion of negative tweets about a visitor posted within an 60 minutes/day leads to lower returns and a higher short book for its stock (even subsequently controlling for traditional-media news sentiment). A higher return for a stock in an hour/day leads to less negative sentiment within the adjacent menses. Nevertheless, we are unable to observe a link between the ii variables when looking at 15-minute intervals. These results are robust to diverse specifications and alternative measures of sentiment. Our findings advise that social media sentiment includes signals beyond those found in traditional media that can bear upon the stock market place. The observed results suggest that these signals are only considered reliable when the sentiment persists over a long enough period.
Access options
Buy single commodity
Instant access to the full article PDF.
34,95 €
Price includes VAT (Indonesia)
Revenue enhancement adding will be finalised during checkout.
Notes
-
For example, if at that place are 6 tweets in a xv-minute period about stock X with sentiment values of -1, -0.five, 0, +0.75, +0.8, and +1.5, the boilerplate sentiment associated with the stock would be 0.26. A similar exercise is carried out when aggregating tweet sentiment to the hourly and daily levels.
-
For example, if there are half dozen tweets in a 15-minute period about stock Ten with sentiment values of -1, -0.five, 0, +0.75, +0.viii, and +1.5, the boilerplate positive sentiment would exist \(\frac{(0.75+0.fourscore+1.50)}{3}=i.02\), whereas the boilerplate negative sentiment would be \(\frac{1.0+0.5}{two}=0.75\).
-
For example, if at that place are half dozen tweets in a fifteen-minute period nearly stock Ten with sentiment values of -ane, -0.5, 0, +0.75, +0.8, and +1.5, the share with positive sentiment would be \(\frac{iii}{6}\), whereas the share with negative sentiment would be \(\frac{2}{six}\).
References
-
Bollen J, Mao H, Zeng 10 (2011) Twitter mood predicts the stock market. Journal of Computational Scientific discipline 2(ane):i–8
-
Checkley MS, Higïœn DA, Alles H (2017) The hasty wisdom of the mob: How market place sentiment predicts stock market place behavior. Adept Systems with Applications 77:256–263
-
Chen H, De P, Hu Y, Hwang B (2014) Wisdom of crowds: the value of stock opinions transmitted through social media. Review of Fiscal Studies 27(5):1367–1403
-
Dougal C, Engelberg J, Garcïœa D, Parsons CA (2012) Journalists and the stock marketplace. Review of Financial Studies 25(three):639–679
-
Ge Q, Kurov A, Wolfe G (2019) Do investors care about presidential company-specific tweets? Journal of Fiscal Research 42:213–242
-
Geweke J (1978) Temporal aggregation in the multiple regression model. Econometrica 46(3):643–661
-
Hu Thou, Liu B (2004) Mining and summarizing client reviews
-
Jiao P, Veiga A, Walther A (2020) Social media, news media, and the stock market. Journal of Economical Behavior & Organisation 176:63–90
-
Kim S, Kim D (2014) Investor sentiment from net message postings and the predictability of stock returns. Journal of Economic Behavior & System 107:708–725
-
Lachanski Yard, Pav S (2017) Shy of the graphic symbol limit: twitter mood predicts the stock market revisited. Econ Periodical Watch xiv(3):302–345
-
Luo X, Zhang J, Duan West (2012) Social media and firm equity value. Information Systems Research 24(1):146–163
-
McGurk Z, Nowak A, Hall JC (2020) Stock returns and investor sentiment: textual analysis and social media. J Econ Finan 44:458–485. https://doi.org/10.1007/s12197-019-09494-iv
-
Mittermayer MA, Knolmayer GF (2006) Newscats: A news categorization and trading system. In: Proceedings of the 6th international conference on information mining, (ICDM), IEEE, pp 1002–1007
-
Nassirtoussi AK, Aghabozorgi S, Ying Wah T, Ling Ngo DC (2014) Text mining for market place prediction: A systematic review. Expert Systems with Applications 41(sixteen):7653–7670
-
Rakowski D, Shirley South, Stark J (2020) Twitter activity, investor attention, and the improvidence of data. Financial Management, 1-44
Acknowledgements
The authors would like to thank the Office of the President of Texas A&M Academy-San Antonio for the Strategic Plan Initiative grant (awarded during the 2018-2019 academic year) that funded this enquiry. Nosotros would also like to thank American Journal Experts (AJE) for English language editing and manuscript formatting. However, the opinions, findings, and conclusions or recommendations expressed in the paper are those of the authors and exercise not necessarily reverberate the views of the Texas A&M Academy-San Antonio
Author information
Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
We have no conflicts of interest to disclose.
Rights and permissions
About this article
Cite this commodity
Fekrazad, A., Harun, S.M. & Sardar, N. Social media sentiment and the stock market. J Econ Finan 46, 397–419 (2022). https://doi.org/ten.1007/s12197-022-09575-ten
-
Accepted:
-
Published:
-
Result Date:
-
DOI : https://doi.org/10.1007/s12197-022-09575-ten
Keywords
- Social media
- Stock market
- Sentiment
- Loughran-McDonald lexicon
- Panel regression
JEL nomenclature:
- C12
- G14
0 Response to "Social Media Sentiment Stock Market"
Post a Comment