Utilizarea momentelor statistice în selecția portofoliilor de criptomonede. O abordare bazată pe analiza cluster
Use of statistical moments in the selection of cryptocurrency portfolios. An approach based on cluster analysis

Summary/Abstract: The paper complements the specialized literature focusing on managing cryptocurrency portfolios based on statistical moments and cluster analysis. We perform the analysis for the first 20 cryptocurrencies, in terms of market capitalization, using their daily closing prices for the entire period of 2020. Thus, our research is structured in three parts, as follows: in the first part of the study it is used the calculation daily returns, for each cryptocurrency, and based on them is highlighted the so-called stylized empirical facts, characteristic of financial time series, after which the focus is exclusively on calculating the first 4 statistical moments of the series of returns, namely the average, the deviation quadratic mean, skewness and kurtosis; the second part is describes application of algorithms for cryptocurrencies groups based on each of the 4 statistical moments calculated previously in order to highlight the common characteristics or of cryptocurrencies in terms of profitability and risk; the last part of the study is reserved to the conclusions and the main future directions of extension of this study.

• Details
• Contents