Twitter-Based uncertainty and cryptocurrency returns

Abstract We explore the relationship between two novel Twitter-based measures of economic and market uncertainty and the performance of four major cryptocurrencies. Using a battery of methods—quantile regressions, Granger-causality in distributions using copula functions, and directional predictability tests—we examine the behavior of Bitcoin, Ethereum, Bitcoin Cash, and Ripple. Our findings demonstrate a strong causal link between the uncertainty expressed in social media and cryptocurrency returns. The effect is particularly evident for Bitcoin and in the tails of return distributions. Our results cast new light on the importance of cryptocurrencies as an alternative asset class in the wake of global uncertainty.