COST FinAI - Fintech and AI in Finance - WG1 Monthly research seminar

Title: FRM for cryptocurrency
Speaker: Anna Shchekina
International Research Training Group 1792, Humboldt-Universität zu Berlin, Germany.

Abstract: In light of the current crisis, cryptocurrencies are drawing attention from even the most conservative investors and promise to set foot in the realm of conventional assets. As a highly connected asset class, any risk measure for it must account for links and mutual dependencies, especially tail events. The FRM Financial Risk Meter is a well-developed technique that - based on Lasso quantile regression - is able to identify individual risk characteristics in a network topology. Hereinafter, we are attempting to explain why FRM, being an average of penalisation parameters, proves to be a good estimate for market risk, focusing particularly on the case of the cryptocurrency market.