METHODOLOGY FOR MAPPING CRYPTOCURRENCY IN THE FINANCIAL MARKET

Abstract

ABSTRACT Cryptocurrencies are traded between consenting parties with no broker and tracked on digital ledgers. On June 7, 2022, two bipartisan U.S. senators introduced a bill, named the Responsible Financial Innovation Act, which classifies cryptocurrency as commodities and gives the Commodity Futures Trading Commission the primary responsibility to oversee the cryptocurrency market. However, Kim et al. (2022) has used a non-Euclidean approach to show that cryptocurrency acts more like securities than commodities according to their price movement. In this paper, we present several non-Euclidean distances, in addition to the Riemannian distance adopted by Kim et al. (2022), which can be used to classify cryptocurrency in the financial market. The computational experiments show that all of the non-Euclidean distances show consistency in the classification analysis to a certain extent. However, their comparison results are different from that of Euclidean distance. For instance, all non-Euclidean distances identify unanimously that shares of large-cap information technology companies are correlated the most with cryptocurrency, but the Euclidean distance suggests that real estate is the most correlated sector of the economy with cryptocurrency. Keywords Commodities, Correlation Matrix, Cryptocurrency, -Means Clustering, Multi-dimensional Scaling, Non-Euclidean Distances, Securities