Nowcasting bitcoin’s crash risk with order imbalance

The spectacular nature of bitcoin price crashes baffles market spectators and prompts routine warnings from regulators cautioning that cryptocurrencies behave in contra to the fundamental properties traditionally define what constitutes money. Arguably most concerning public is, first, bitcoin’s unprecedented volatility relative other asset classes and, second, its seemingly detached behavior time-honored economic fundamentals. In an attempt create early warning system crash risk using generalized extreme value (GEV) logistic regression modeling, this study integrates order flow imbalance, along with several control factors which reflect blockchain activity network value, nowcast crashes. From a data analysis perspective, despite their dissimilar distributional underpinnings, GEV models perform comparably. When evaluating type I II errors these yield, it is shown performance comparable terms accuracy. addition, also how proportion can shift dramatically across probability cutoff tolerances. Towards end study, time varying probabilities are evaluated. sample range encompasses SARS-CoV-2 (Covid-19) period as well recent scandal collapse FTX cryptocurrency exchange.