Title: Application of machine learning methods to forecast cryptocurrencies volatility Authors:  Witold Orzeszko - Nicolaus Copernicus University in Torun (Poland) [presenting]
Piotr Fiszeder - Nicolaus Copernicus University in Torun (Poland)
Grzegorz Dudek - Czestochowa University of Technology (Poland)
Pawel Kobus - Warsaw University of Life Sciences (Poland)
Abstract: A comprehensive study of statistical and machine learning methods is presented for predicting the volatility of the following four cryptocurrencies: Bitcoin, Ethereum, Litecoin, and Monero. Several methods, i.e., HAR, ARFIMA, GARCH, LASSO, ridge regression, SVR, MLP, fuzzy neighbourhood model, random forest, and LSTM, are compared in terms of their forecasting accuracy. The realized variance calculated from intraday returns is used as the input variable for the models. The experimental results demonstrate that there is no single best method for forecasting the volatility of each cryptocurrency, and different models may perform better depending on the specific cryptocurrency, choice of the error metric and forecast horizon. Furthermore, it is shown that simple linear models such as HAR and ridge regression do not perform worse than more complex models like LSTM and random forest. The research provides a useful reference point for the development of more complex models and suggests the potential benefits of incorporating additional input variables.