A literature review on the predictability of volatility in cryptocurrencies

Abstract

In recent years, cryptocurrency has grown in importance and weight, in financial markets, with Bitcoin being the first decentralized peer-to-peer payment network and altcoins enhancing Bitcoin's speed and privacy. With the initial coin offerings in 2016, Ethereum came to gain market in comparison to Bitcoin. The technology underlying cryptocurrencies is revolutionary, with blockchain technology having been extensively researched in several areas and classified into three basic types: public, private, and federated/consortium blockchain, according to Zheng et al. (2018). In this dissertation, a systematic literature review about the methods used to forecast various cryptocurrencies price and return volatility is done. The SLR (Systematic Literature Review) technique and data analytics approach were used to analyze the publications from Elsevier, Emerald, and Google Scholar. Text mining tools were utilized to analyze the articles, which were run through R software. The output revealed that the GARCH model was the most used in the selection literature, followed by the Wavelet-Base DCC, and Machine-Learning methods. Various authors presented many GARCH-type models to explain and estimate the price and return volatility of Bitcoin and other cryptocurrencies in order to present which models type can better explain the cryptocurrencies volatility. Though, to present more accurate GARCH models that predict the volatility of cryptocurrencies, there are authors who go beyond the traditional GARCH and investigate hybrid GARCH models with machine learning techniques.

Date of Award	18 Dec 2023
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Gonçalo Faria (Supervisor) & Fabio Verona (Co-Supervisor)