Analysis of the cryptocurrency market using different prototype-based clustering techniques
Abstract and Figures
Since the emergence of Bitcoin
, cryptocurrencies have grownup significantly, not solely in terms of capitalization however additionally in number. Consequently, the cryptocurrency market may be a contributive arena for investors, because it offers several opportunities. However, it's tough to understand. This study aims to describe, summarize, and phase the most trends of the complete cryptocurrency market in 2018, exploitation information analysis tools. Accordingly, we tend to propose a brand new bunch-based methodology that gives complementary views of the monetary behavior of cryptocurrencies, and one that appears for associations between the clustering results, and different factors that don't seem to be concerned in clustering. Particularly, the methodology involves applying 3 different partitional bunch algorithms, wherever every of them use a unique illustration for cryptocurrencies, namely, yearly mean, and variance of the returns, distribution of returns that haven't been applied to monetary markets previously, {and the |and therefore the |and additionally the} statistic of returns. as a result of each representation provides a different outlook of the market, we tend to also examine the combination of the three clustering results, to get a fine-grained analysis of the most trends of the market. In conclusion, we analyze the association of the clustering results with different descriptive options of cryptocurrencies,
together with the age, technological attributes, and financial ratios derived from them. this can facilitate to boost the identification of the clusters with further descriptive insights, and to seek out associations with different variables. Consequently, this study describes the full market supported graphical information, and a climbable methodology that may be reproduced by investors who wish to grasp the most trends within the market quickly, and people that search for cryptocurrencies with different monetary performance .In our analysis of the 2018 and 2019 for extended period, we tend to found that the market may be usually segmental in few clusters (five or less), and even considering the intersections, the half-dozen more populations account for 75% of the market. relating to the associations between the clusters and descriptive features, we discover associations between some clusters with volume, market capitalization, and a few monetary ratios, that can be explored in future research.
