2018/05/01 by Tianyu Ray Li, Li, Tianyu Ray, Anup S. Chamrajnagar +7
Computer Science · Economics, Econometrics and Finance · #Blockchain Technology Applications and Security #Complex Systems and Time Series Analysis #Computational Finance (q-fin.CP) #FOS: Economics and business #FOS: Physical sciences #Physics and Society (physics.soc-ph)
paper · pdf · doi:10.48550/arxiv.1805.00558
openalex publication_date 2018/05/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this paper, we analyze Twitter signals as a medium for user sentiment to\npredict the price fluctuations of a small-cap alternative cryptocurrency called\n\ZClassic. We extracted tweets on an hourly basis for a period of 3.5\nweeks, classifying each tweet as positive, neutral, or negative. We then\ncompiled these tweets into an hourly sentiment index, creating an unweighted\nand weighted index, with the latter giving larger weight to retweets. These two\nindices, alongside the raw summations of positive, negative, and neutral\nsentiment were juxtaposed to \∼ 400 data points of hourly pricing data to\ntrain an Extreme Gradient Boosting Regression Tree Model. Price predictions\nproduced from this model were compared to historical price data, with the\nresulting predictions having a 0.81 correlation with the testing data. Our\nmodel's predictive data yielded statistical significance at the p < 0.0001\nlevel. Our model is the first academic proof of concept that social media\nplatforms such as Twitter can serve as powerful social signals for predicting\nprice movements in the highly speculative alternative cryptocurrency, or\n"alt-coin", market.\n