Sentiment, Google queries and explosivity in the cryptocurrency market
a year ago
Sentiment, Google queries and explosivity in the cryptocurrency market

A new article by the University of Pavia (UNIPV) has been launched in Physica A: Statistical Mechanics and its Applications: “Sentiment, Google queries and explosivity in the cryptocurrency market”



The lack of fundamental values in the cryptocurrency market paves the way for the rise of unprecedented speculative bubble phenomena, which are often associated with alternating phases of investors’ fear and greed. We propose exploiting the information derived from a large set of cryptocurrency news and Google Search Indices to detect and, possibly, anticipate the presence of speculative bubbles in cryptocurrency prices. This is done using a Backward Superior Covariate-Augmented Dickey–Fuller (BSCADF) test, which allows us to explicitly account for market sentiment when testing the presence of an explosive root in cryptocurrency prices. Our results show that the covariate test statistics does significantly diverge from the traditional statistics in concomitance with price surges, highlighting the ability of sentiment to predict speculative bubble occurrences. We also show how a polarised version of investors’ sentiment plays an overall more determinant role, if compared to news volume and Google queries, in providing an early warning signal of market bubble episodes in cryptocurrencies.


Find the full article here!