A comparison of pricing models for cryptocurrencies

Abstract

For recent years, thousands of cryptocurrencies have been released and contributed to a large proportion of trading and investing activity. However, there are immensely limited and contagious pricing models for this type of digital asset since the cryptocurrency market could include many novel risk elements. In this research, the strategy 1-1 for momentum confirmation and multiple regression for time-series data are used to compare the explanatory power of four common stock pricing models: CAPM, Shen et al three-factor model, Fama & French three-factor model and Carhart four-factor models on 200 largest cryptocurrencies in the period 2015-2020. The results indicate the momentum and size effect on the cryptocurrency market. Regarding regression analysis, the Carhart model obtains the highest explanatory power while CAPM denotes the lowest one. The study also detects the nearly similar interpreting capacity of two remaining three-factor pricing models in which momentum factor replaces for value factor. Keywords: cryptocurrency, pricing model, Carhart four-factor model, Fama & French three-factor model, Shen et al. (2019) three-factor model, CAPM