Analysing Social Media Forums to Discover Potential Causes of Phasic Shifts in Cryptocurrency Price Series

The recent extreme volatility in cryptocurrency prices occurred in the setting of popular social media forums devoted to the discussion of cryptocurrencies. We develop a framework that discovers potential causes of phasic shifts in the price movement captured by social media discussions. This draws on principles developed in healthcare epidemiology where, similarly, only observational data are available. Such causes may have a major, one-off effect, or recurring effects on the trend in the price series. We find a one-off effect of regulatory bans on bitcoin, the repeated effects of rival innovations on ether and the influence of technical traders, captured through discussion of market price, on both cryptocurrencies. The results for Bitcoin differ from Ethereum, which is consistent with the observed differences in the timing of the highest price and the price phases. This framework could be applied to a wide range of cryptocurrency price series where there exists a relevant social media text source. Identified causes with a recurring effect may have value in predictive modelling, whilst one-off causes may provide insight into unpredictable black swan events that can have a major impact on a system.

[1]  Peter A. Gloor,et al.  Nowcasting the Bitcoin Market with Twitter Signals , 2014, ArXiv.

[2]  Young Bin Kim,et al.  Predicting Fluctuations in Cryptocurrency Transactions Based on User Comments and Replies , 2016, PloS one.

[3]  Ladislav Kristoufek,et al.  BitCoin meets Google Trends and Wikipedia: Quantifying the relationship between phenomena of the Internet era , 2013, Scientific Reports.

[4]  F. Dablander,et al.  Node centrality measures are a poor substitute for causal inference , 2018, Scientific Reports.

[5]  Andrew Burnie,et al.  Developing a Cryptocurrency Assessment Framework: Function over Form , 2018, Ledger.

[6]  Michel Terraza,et al.  Testing for Causality , 1994 .

[7]  Andrea Baronchelli,et al.  Wikipedia and Cryptocurrencies: Interplay Between Collective Attention and Market Performance , 2019, Front. Blockchain.

[8]  Paul R. Rosenbaum,et al.  Observation and Experiment: An Introduction to Causal Inference , 2017 .

[9]  R. Shiller.  Irrational Exuberance Ed. 2 , 2005 .

[10]  John Nelson,et al.  Cryptocurrency Price Prediction Using Tweet Volumes and Sentiment Analysis , 2018 .

[11]  J. Pearl,et al.  Causal Inference in Statistics: A Primer , 2016 .

[12]  Jaegul Choo,et al.  When Bitcoin encounters information in an online forum: Using text mining to analyse user opinions and predict value fluctuation , 2017, PloS one.

[13]  Xin Li,et al.  The Technology and Economic Determinants of Cryptocurrency Exchange Rates: The Case of Bitcoin , 2014, Decis. Support Syst..

[14]  Emine Yilmaz,et al.  An Analysis of the Change in Discussions on Social Media with Bitcoin Price , 2019, SIGIR.

[15]  Mark Cummins,et al.  Psychological Barriers in Oil Futures Markets , 2013 .

[16]  A. B. Hill.  The Environment and Disease: Association or Causation? , 1965, Proceedings of the Royal Society of Medicine.

[17]  LiXin,et al.  The technology and economic determinants of cryptocurrency exchange rates: The case of Bitcoin , 2017, Decis. Support Syst..

[18]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[19]  Frank Schweitzer,et al.  Social signals and algorithmic trading of Bitcoin , 2015, Royal Society Open Science.

[20]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[21]  K. Schulz,et al.  Bias and causal associations in observational research , 2002, The Lancet.

[22]  P. Bridge,et al.  Increasing physicians' awareness of the impact of statistics on research outcomes: comparative power of the t-test and and Wilcoxon Rank-Sum test in small samples applied research. , 1999, Journal of clinical epidemiology.

[23]  M. Maathuis,et al.  Estimating high-dimensional intervention effects from observational data , 2008, 0810.4214.

[24]  Frank Westerhoff,et al.  Anchoring and Psychological Barriers in Foreign Exchange Markets , 2003 .

[25]  Andrew Burnie,et al.  Social media and bitcoin metrics: which words matter , 2019, Royal Society Open Science.

[26]  Geoffrey Lightfoot,et al.  Price Fluctuations and the Use of Bitcoin: An Empirical Inquiry , 2015, Int. J. Electron. Commer..

[27]  Raj Aggarwal,et al.  Psychological Barriers in Gold Prices? , 2005 .

[28]  Michele Marchesi,et al.  Bitcoin Spread Prediction Using Social and Web Search Media , 2015, UMAP Workshops.

[29]  Frode Kjærland,et al.  How can Bitcoin Price Fluctuations be Explained , 2018 .

[30]  Pavlin Mavrodiev,et al.  The digital traces of bubbles: feedback cycles between socio-economic signals in the Bitcoin economy , 2014, Journal of The Royal Society Interface.

[31]  John P A Ioannidis,et al.  Exposure‐wide epidemiology: revisiting Bradford Hill , 2016, Statistics in medicine.

[32]  C. Granger.  Testing for causality: a personal viewpoint , 1980 .

[33]  Scott Henderson,et al.  Pathology, Causes, Course and Treatment of Rheumatism , 1873, The Chicago medical journal.

[34]  Aleh Tsyvinski,et al.  Risks and Returns of Cryptocurrency , 2018, The Review of Financial Studies.

[35]  E. C. Hammond,et al.  Smoking and lung cancer: recent evidence and a discussion of some questions. , 1959, Journal of the National Cancer Institute.

[36]  Andrew Urquhart,et al.  What causes the attention of Bitcoin , 2018 .

[37]  Stephen Kokoska,et al.  Chance Encounters: A First Course in Data Analysis and Inference , 2001, Technometrics.

[38]  Alexander M. Millkey.  The Black Swan: The Impact of the Highly Improbable , 2009 .

[39]  C. Herff,et al.  Predicting altcoin returns using social media , 2018, PloS one.

[40]  Rick Wash,et al.  How "True Bitcoiners" Work on Reddit to Maintain Bitcoin , 2019, CHI Extended Abstracts.

[41]  A. B. Hill,et al.  "The Environment and Disease: Association or Causation?" (1965), by Austin Bradford Hill , 2017 .

[42]  Bernhard Schölkopf,et al.  Inferring causation from time series in Earth system sciences , 2019, Nature Communications.