Abstract

Abstract

This paper provides an overview of methods for testing the predictability of macroeconomic indicators based on real-time data revisions. Various approaches to their forecasting are considered and their properties are discussed.

Abstract

An automatic method is proposed for the removal of the initialization bias that is intrinsic to the output of any statistically stationary simulation. The general techniques based on optimization approaches such as Beyhaghi et al. [1] following the Marginal Standard Error Rules (MSER) method of White et al. [16] were observed to be highly sensitive to the fluctuations in a time series and resulted in frequent overprediction of the length of the initial truncation. As fluctuations are an innate part of turbulence data, these techniques performed poorly on turbulence quantities, meaning that the local minima was often wrongly interpreted as the minimum variance in the time series and resulted in different transient point predictions for any increments to the sample size. This limitation was overcome by considering the finite difference of the slope of the variance computed in the optimization algorithm. The start of the zero slope region was considered as the initial transient truncation point. This modification to the standard approach eliminated the sensitivity of the scheme, and led to consistent estimates of the transient truncation point, provided that the finite difference time interval was chosen large enough to cover the fluctuations in the time series. Therefore, the step size for the finite difference slope was computed using both visual inspection of the time series and trial and error. We propose the Augmented Dickey­Fuller test as an automatic and reliable method to determine the truncation point, from which the time series is considered stationary and without an initialization bias.

Abstract

More than 12 years have passed since the debut of Bitcoin, the first cryptocurrency, but there is still no clear understanding of the essence of cryptocurrencies. At the same time, in recent years, interest in cryptocurrencies has continued to grow, and this financial instrument is becoming more and more attractive to individuals. Therefore, an issue of essence of cryptocurrencies is relevant. The main subject of the study is the profitability of cryptocurrencies. The main aim of this work is to identify the determinants of cryptocurrency returns. To achieve this goal, such tasks as the creation of factors reflecting the characteristics of the cryptocurrency market and the use of multifactor Fama-French models for the analysis of cryptocurrency returns were performed. Based on the collected daily data on capitalization, trading volumes and the price of more than 5,000 cryptocurrencies for the period from 01.04.2014 to 21.06.2021, standard factors based on capitalization indicators, trading volumes and the third momentum were built. The main estimation method is econometric modeling using the least squares method. The obtained results of an empirical study indicate a positive relationship between the profitability of groups of cryptocurrencies and the difference in the yields of the upper and lower 30% of cryptocurrencies according to the third moment. The difference in the yields of the lower and upper 30% of cryptocurrencies by market capitalization has a negative impact on the profitability of groups of cryptocurrencies. The main conclusion of the study is that before the beginning of high volatility period, cryptocurrencies could be considered as an asset for the diversification of market risk, but subsequently the cryptocurrency market began to move co-directionally to the stock market. The scientific novelty of the work stems from presenting an assessment of the impact of modeled factors on various groups (portfolios) of cryptocurrencies over certain periods of time. The study recommends to conduct further analysis of the profitability factors of cryptocurrencies on more homogeneous samples.

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.

Abstract

Abstract

Abstract