Correlation analysis of the most popular crypto-currencies and their price prediction

Vekios, Panagiotis; Βέκιος, Παναγιώτης

dc.contributor.author	Vekios, Panagiotis	en
dc.contributor.author	Βέκιος, Παναγιώτης	el
dc.date.accessioned	2024-01-26T08:35:51Z
dc.date.available	2024-01-26T08:35:51Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/58658
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.26354
dc.rights	Default License
dc.subject	Cryptocurrency	en
dc.subject	Κρυπτονόμισμα	el
dc.subject	Correlation analysis	en
dc.subject	Price prediction	en
dc.subject	Machine learning	en
dc.subject	Ανάλυση συσχέτισης	el
dc.subject	Πρόβλεψη τιμών	el
dc.subject	Μηχανική μάθηση	el
dc.subject	Bitcoin	en
dc.title	Correlation analysis of the most popular crypto-currencies and their price prediction	en
heal.type	bachelorThesis
heal.classification	Data Analytics & Machine Learning	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2023-06-20
heal.abstract	The objective of this thesis is to investigate the correlation between the most popular cryptocurrencies and explore their potential for price prediction. The rapid growth and widespread adoption of cryptocurrencies have attracted significant attention from investors and researchers alike. Understanding the interrelationships between different cryptocurrencies can provide valuable insights into the dynamics of this emerging market. To conduct this study, a comprehensive dataset comprising historical price and trading volume data of the most popular cryptocurrencies was collected and analyzed. The selected cryptocurrencies include Bitcoin (BTC), Ethereum (ETH), and Litecoin (LTC), which are known for their dominant market positions and wide-ranging impact on the cryptocurrency ecosystem. The first phase of the research involves analyzing the correlation between the selected cryptocurrencies, but also the evolution of their route since the day of their creation. The correlation analysis aims to quantify the strength and direction of relationships between these digital assets. By examining correlations, it can be determined whether the prices of different cryptocurrencies move in sync or independently from one another. This analysis will shed light on potential diversification benefits and the degree of interdependence within the cryptocurrency market. Furthermore, this thesis explores the use of correlation analysis as a tool for predicting cryptocurrency prices. By leveraging historical correlation patterns, it can be assessed whether the correlation between certain cryptocurrencies can be indicative of future price movements. Statistical models and machine learning algorithms will be employed to identify and quantify the predictive power of correlations in forecasting cryptocurrency prices. The results of this study will contribute to the existing body of knowledge on cryptocurrencies and provide valuable insights for investors, traders, and financial analysts. A better understanding of the correlation patterns and their potential predictive abilities can assist in making informed investment decisions and developing effective trading strategies in the volatile cryptocurrency market. Finally, this thesis explores the utilization of machine learning techniques for price prediction of cryptocurrencies. By leveraging advanced machine learning models, models can be developed that exploit the characteristics of cryptocurrencies and their correlations to more accurately forecast future prices. Keywords: cryptocurrency, correlation analysis, price prediction, machine learning, Bitcoin, Ethereum	en
heal.sponsor	This thesis was devised in the Computer Laboratory Systems of the field of Information Technology and Computers of the School of Electrical and Computer Engineering NTUA. On the occasion of the completion of my thesis, I would like to express my warm thanks to my supervisor, Professor Mr. Nectarios Koziris, for the opportunity he gave me to work in research in the Blochchain, Data and Machine Learning fields, but also for the influence he exerted on me from classes throughout my time at this school. Also, I would like to especially thank Dr. Katerina Doka for her supervision and valuable contribution to the preparation of this diplomatic work. The directions she provided me and the constant interest she showed were crucial to its completion. In closing, I would like to thank the members of the examination committee Mr. Nectarios Koziris, Mr. Ioannis Konstantinou, and Mr. Dimitrios Tsoumakos for the time they will devote to the study of my thesis. Finally, I would like to express my gratitude to my family and friends who have shown me unconditional support during my studies.	en
heal.advisorName	Koziris, Nectarios	en
heal.committeeMemberName	Konstantinou, Ioannis	en
heal.committeeMemberName	Tsoumakos, Dimitrios	en
heal.committeeMemberName	Koziris, Nectarios	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών. Τομέας Τεχνολογίας Πληροφορικής και Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	107 σ.	el
heal.fullTextAvailability	false
heal.fullTextAvailability	false