Title: Testing asset pricing models on the cryptocurrency market
Authors: Francesco Violante - ENSAE ParisTech (France) [presenting]
Stefano Grassi - University of Rome 'Tor Vergata' (Italy)
Abstract: The purpose is to test three and a five-factor models capturing the size, value, momentum and short term reversal in the cross-section of cryptocurrencies returns to examine if these are sufficient to capture market-wide sources of risk. Adapted versions of the Fama-French SMB and HML long-short value-weighted portfolios meaningful to the cryptocurrency market are created using the universe of cryptocurrencies available for the period 2015 to 2019. Inspired by the Cholesky GARCH model, we develop a multivariate conditional beta model based on a block LDU decomposition of the conditional covariance matrix of the system including factors and asset returns. The model allows estimating in a single pass the asset-specific exposure to the risk factors and the premium associated with such exposure. The asset pricing models are tested on the cross-section of 25 value-weighted portfolio sorts by size and network value to transactions ratio.