Forecasting Bitcoin volatility using machine learning techniques
比较了LSTM、CNN-LSTM等机器学习模型与传统GARCH、HAR模型在比特币波动率预测上的表现，发现神经网络在1天至2个月的所有预测期上均优于GARCH，且CNN-LSTM结合马尔可夫转移场在7天短期预测上效果最佳。
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 Neural Network-LSTM (CNN-LSTM) model to the traditional models. We find that neural networks outperform Generalised Autoregressive Conditional Heteroskedasticity (GARCH) models for all forecasting horizons. Furthermore, the LSTM model outperforms the Heterogeneous Autoregressive (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. • We forecast Bitcoin volatility using intraday data with machine learning models. • High-frequency Bitcoin data benefits Bitcoin volatility predictions. • We convert time series to images to improve Bitcoin volatility prediction. • Our approach outperforms HAR and GARCH, especially in short-term forecasts. • Image transformation can capture non-linear features such as clustering effect.