A0417
Title: Blockchain characteristics and systematic risk: A neural network based factor model for cryptocurrencies
Authors: Alla Petukhina - HTW Berlin (Germany) [presenting]
Abstract: A neural network-based factor model is applied to the cryptocurrency market to describe individual asset returns in terms of latent risk factors and time-varying risk exposures. Five contributions are made. First, it is shown that pricing performance is improved by adding nonlinearities to the risk exposures. Second, it is established that the risk dynamics of the cryptocurrency market evolve more quickly than for equity. Third, it is identified that cryptocurrency prices were more predictable before the COVID-19 pandemic than thereafter. Fourth, latent risk factors are found to be related to observables but additionally include idiosyncratic variance. Last, it is observed that asset characteristics that are important for the estimation of risk exposures are commonly found in the literature on observable factor models.
