Multiscale Multifractal Detrended Analysis of Speculative Attacks Dynamics in Cryptocurrencies

Abstract

Cryptocurrencies have drawn the interest of both scholars and professionals due to their decentralised, unique payment system supported by blockchain technology and their autonomy from sovereign governments, centralised organisations, and banking systems. Numerous works have studied, on the one hand, the behavior of cryptocurrencies, and on the other hand, the multifractal model in financial markets. Nevertheless, the limitations of existing models exist, and the literature calls for more research into multifractal analysis techniques applied to finance, as the methodology widely used in previous research is the regression model and machine learning methods. This study introduces a new model for predicting unexpected situations of speculative attacks in the cryptocurrency market, applying the method of Multiscale Multifractal Detrended Fluctuation Analysis, which shows excellent precision results. Our approach has a high impact potential on the forecast of possible speculative actions over the value of cryptocurrencies and against the risks derived from the control of cryptocurrencies by private entities, so the question of the possible effect on the financial system is of great importance.