Title: A factor model for cryptocurrency returns
Authors: Daniele Bianchi - Queen Mary University of London (United Kingdom) [presenting]
Mykola Babiak - Lancaster University Management School (United Kingdom)
Abstract: The factor structure of a large cross-section of daily returns on digital assets is investigated through the lens of an Instrumented Principal Component Analysis (IPCA). We show that a model with three latent factors and time-varying factor loadings significantly outperforms a benchmark six-factor model with traded, observable, factors: the in-sample (out-of-sample) $R^2$ from the IPCA stands at 17\% (2.8\%) for daily returns, against a benchmark 8\% (0.2\%) obtained from a model with observable risk factors taken from the existing literature. By looking at the characteristics that significantly matter for the dynamics of the latent factors, we provide robust evidence that risk premiums for digital assets are primarily driven by liquidity, volatility, downside risk, and the market beta. The results hold both from a sample of daily returns from December 2nd 2016 to July 9th 2021 and across different sub-samples. Due to the inherent differences in cryptocurrencies and the novel and emergent status of digital assets as a form of investment, this research will be relevant to a broad audience; from market participants seeking different sources of returns and diversification, to regulators wishing to understand the role of digital assets, and to academics searching for new insights into the drivers and risks in cryptocurrency markets.