THE METHODOLOGY FOR SHORT-TERM FORECASTING OF CRYPTOCURRENCY EXCHANGE RATES

Abstract

The purpose of this article is to present the methodology of short-term forecasting of cryptocurrency exchange rates, which describes the main stages of statistical research using methods of intelligent data analysis and machine learning. The growing number of cryptocurrencies and their high volatility have made cryptocurrency exchange rate forecasting an important research topic. Short-term forecasting of cryptocurrency exchange rates is difficult due to their high volatility, lack of a fundamental valuation model, and the influence of many factors. Creating a methodology for short-term forecasting of cryptocurrency exchange rates provides a comprehensive basis for further statistical analysis and constructing a forecasting model. Methodology of research. Research methodology is the collection and analysis of historical data using the following methods: system analysis, analytical method (statistical analysis). Findings. The scheme of the CRISP-DM standard has been adapted for further research on the forecasting of cryptocurrency exchange rates. The framework consists of 8 modules covering all the necessary steps in the forecasting process, from data understanding and pre-processing to model selection, training, evaluation and deployment. Practical value. Developing a methodology for short-term forecasting requires extensive data analysis and research. This practical value lies in advancing statistical techniques, machine learning algorithms, and intelligent data analysis methods. It contributes to the overall understanding of cryptocurrency markets and facilitates further research in the field. The results of this study have practical implications for investors and risk managers looking for tools to predict short-term cryptocurrency prices. The proposed methodology can be used to make investment decisions and improve risk management strategies by providing accurate forecasts of future values of cryptocurrency exchange rates. This can help investors make more informed decisions about buying and selling cryptocurrencies, and help risk managers develop strategies to reduce the resulting costs.