Abstract

There is still no common understanding of whether cryptocurrencies should be classified as financial assets or as currencies. The ambiguity and versatility of the definition of the nature and functions of cryptocurrencies give rise to a variety of views on the methods of modeling their returns. Therefore, the issue of essence of cryptocurrencies is topical. The main subject of the study is the return of cryptocurrencies. The main aim of this work is to identify the determinants of return of homogeneous groups of cryptocurrencies. To achieve this goal, such tasks as the formation of various groups of cryptocurrencies, modeling of factors that take into account the peculiarities of the cryptocurrency market, and the evaluation of multifactor models of the Fama-French type for the analysis of cryptocurrency returns have been performed. Based on the collected daily data on capitalization, trading volumes and the price of cryptocurrencies for the period from 01.04.2014 to 29.05.2022, standard factors for cryptocurrencies based on market capitalization, trading volumes and the first momentum, as well as factors reflecting the return of the cryptocurrency market as a whole and the return of the stock market (S&P500) were constructed. The main method of estimating regressions is econometric modeling using the least squares method. The results of an empirical study indicate a positive relationship between the return of homogeneous groups of cryptocurrencies and the difference in the yields of the upper and lower 30% of cryptocurrencies in terms of market capitalization. Weighted return of the cryptocurrency market based on market capitalization (analogous to the S&P500) has a positive impact on the return of homogeneous groups of cryptocurrencies. The main conclusion of the study is that the transition to empirical analysis based on homogeneous groups of cryptocurrencies allowed us to obtain stable results indicating the absence of a relationship between the return of financial assets and the return of cryptocurrencies that are in a single homogeneous group. The scientific novelty of the work consists in presenting an assessment of the impact of modeled factors on various groups (portfolios) of cryptocurrencies in certain periods of time. This study recommends conducting a search for the determinants of cryptocurrency returns and subsequent analysis of their impact.