An Application of Autoregressive Extreme Value Theory to Cryptocurrencies

We study the tails' behavior of four major Cryptocurrencies (Bitcoin, Litecoin, Ethereum and Ripple) by employing the Autoregressive Fŕechet model for conditional maxima. Using five-minute-high-frequency data, we report time-evolving tails as well as provide a straightforward measure of tails asymmetry for positive and negative intra-day returns. We find that only Bitcoin has a notable more massive tail for positive returns asymmetry while the remaining three Cryptocurrencies have a general tendency towards more massive negative intra-day tails. All considered Cryptocurrencies depict lighter tails as the market matures.