Measures of risk in cryptocurrency market using extreme value theory

Abstract

This paper attempts to compute the risk measures in cryptocurrency market using extreme value theory. The extreme market value theory focuses on extreme and rare events that happened in each cryptocurrency in order to calculate the extreme risk measures such as the Value at Risk and the Expected Shortfall. In this paper, the extreme value theory was successfully applied to the cryptocurrency market log returns, and the VaR and ES was successfully predicted. The study was done on seven cryptocurrencies from August 2014 to March 2019. The major findings of this paper are 1-) The VaR, ES and other risk measures estimations using historical simulation, show that Monero, Bitshares, Stellar and Ripple have the highest VaR and the highest ES at 1% and 5% significance. 2-) The extreme value theory was successfully applied and showed that Stellar, Ripple, and Bitshares enclosed the highest VaR and ES at 95% and 99% confidence interval.