Forecasting Bitcoin volatility using machine learning techniques

Huang, Z.-C., Sangiorgi, I. ORCID: https://orcid.org/0000-0002-8344-9983 and Urquhart, A. ORCID: https://orcid.org/0000-0001-8834-4243 (2024) Forecasting Bitcoin volatility using machine learning techniques. Journal of International Financial Markets, Institutions and Money, 97. 102064. ISSN 1873-0612 doi: 10.1016/j.intfin.2024.102064

Abstract/Summary

This paper studies the Bitcoin volatility forecasting performance between popular traditional econometric models and machine learning techniques. We compare the 1-day to 2-month ahead forecasting performance of the Long Short-Term Memory (LSTM) and a hybrid Convolutional Neu- ral Network-LSTM (CNN-LSTM) model to the traditional models. We find that neural networks outperform Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models for all forecasting horizons. Furthermore, the LSTM model outperforms the Heterogeneous Autoregres- sive (HAR) model and by integrating the Markov Transition Field (MTF) into the CNN-LSTM model, we achieve superior forecasting results in the short-term, particularly for the 7-day forecasts.