Can a Hurst exponent be useful for investors in crypto-currency markets?

Oleksii Kryvchun

Bakalářská práce

Abstract:

This bachelor's thesis deals with the revision of the efficient market hypothesis as well as the choice of an appropriate model for analyzing the time series of cryptocurrencies, in particular Bitcoin. The chosen analysis will determine if Bitcoin is a similar type of financial asset and follows the same market rules and patterns. The practical part of the dissertation presents the problem of modern time series analysis that investors face. This bachelor thesis deals with procedures of time series analysis using a main statistical measure of long-term memory of a time series in Fractal market hypothesis – Hurst exponent. Hurst exponent - a modern measure used in time series analysis will be used in the analysis of Bitcoin quotes. This value decreases as the delay between two identical pairs of values in the time series increases. Calculating the Hurst exponent we measure the amount by which a given time series deviates from a random walk. In other words, the Hurst indicator makes it clear if the selected time series maintains the existing trend, the trend tends to change to the opposite, or it has a random movement that cannot be predicted. Based on all the Hurst exponent calculations carried out in Excel, a conclusion will be made regarding the behavior of Bitcoin prices. The possibility of applying the chosen method for the analysis of the crypto market is also being considered.

Keywords

Internet, Bitcoin, Hurst exponent, Brownian motion, Long-term memory, Rescaled range analysis, Money

Jazyk práce: angličtina

Datum vytvoření / odevzdání či podání práce: 31. 3. 2022

Obhajoba závěrečné práce

• Vedoucí: Ing. Pavel Srbek, Ph.D.

Česká zemědělská univerzita v Praze

Provozně ekonomická fakulta

Bakalářský studijní program / obor:
Economics and Management / Economics and Management