Factor Models of Cryptocurrency Returns: Financial Theory Approach

Abstract

The purpose of the article is to identify the determinants of cryptocurrency returns. To achieve this goal, the article presents an attempt to create factors that reflect the characteristics of the cryptocurrency market, and uses Fama–French type multifactor models for analyzing the returns of cryptocurrencies. Standard factors based on capitalization indicators, cryptocurrency trading volumes and the third momentum were built. The paper also presents an estimation of the impact of these factors on various groups, or portfolios, of cryptocurrencies in certain periods of time (the period of market formation and the period of high price volatility of the market, including its division into two sub-periods: before the coronavirus pandemic and during the pandemic), which allows us to consider the heterogeneity of data both in time and for certain indicators. As a result of estimating regressions on daily data, empirical evidence in favor of a positive relationship between the excess return of cryptocurrency groups with the constructed factors was obtained. In addition, the paper checks the relationship between the cryptocurrency market and the stock market. Prior to the beginning of high volatility period, cryptocurrencies could be considered as an asset for the diversification of market risk, but later there could be found co-movement of the cryptocurrency market and the stock market, seen from the appearance of the statistical significance of the coefficient before a variable reflecting the market risk premium. In addition, it was shown that the frequency of data can affect the estimates of the coefficients but does not affect the fundamental conclusions of the analysis. The findings indicate the need for further analysis of the cryptocurrency return factors on more homogeneous samples.