Bitcoin as an investment : predictions and factors affecting it

• Κύριος συγγραφέας: Πλατή, Ευπραξία
• Άλλοι συγγραφείς: Plati, Efpraxia
• Γλώσσα: Greek
• Έκδοση: 2022
• Θέματα: Bitcoin; Investments; Cryptocurrency; Time series analysis; Prediction; Επενδύσεις; Κρυπτονομίσματα; Ανάλυση χρονοσειρών; Πρόβλεψη

Data analysis, that is, the thorough study of a set of information, aims to extract conclusions that could help a company or an entity make better decisions. The time series analysis is a process that has been established in recent years and is used in statistics, pattern recognition, economics, weather forecasting and seismic prediction, astronomy, communications engineering, and generally speaking, in any field of applied sciences and mechanics involving time measurements. Its purpose is to understand the behavior of time series and to detect recurring patterns which in turn lead to useful conclusions and predictive models. In recent years, cryptocurrencies have dominated the media and promise quick and huge profits, making it easy to understand why people succumb to investing large amounts of money, even though these currencies are relatively new. A very promising cryptocurrency is that of the Bitcoin. This new form of «virtual» money has not only been discussed a lot by investors and entrepreneurs but also by the general public. Since Bitcoin prices are collected and recorded daily, they are a remarkable subject in which we can apply time series analysis. The purpose of this research is to investigate the factors that influence the price of the cryptocurrency Bitcoin. In more detail, the data used was taken from the website yahoo finance (Yahoo Finance - Stock Market Live, Quotes, Business Finance News, n.d.) and is concerned with monthly values of the variables for the period Octo- ber 2014 to April 2022. The factors chosen were the German, US, and Japanese stock market indices, as their relationship with cryptocurrency prices have been established (Guizani Nafti, 2019; Mohd Thas Thaker Ah Mand, 2021). Also, pre-eminent investment products, gold, oil, as well as an agricultural investment product, specifically, that of coffee, were investigated. What is more, the forecast was made by applying the Auto-Regressive Integrated Moving Average (ARIMA) method. In the first chapter, the concept of forecasting, as well as the categories of forecasting methods, are analyzed. The second chapter analyzes the Time Series with the definition, characteristics of time series, components, and models of time series. Next chapter, refers to the basic properties of time series, and some forecasting methods are described. Finally, an introduction is made to cryptocurrencies and more specifically, to the Bitcoin, whose data is implemented analyzed and predicted. In other words, the definitions mentioned above will be better understood through time series analysis for the data of Bitcoin, with the aim to forecast and to find the factors that influence it.