Bitcoin volatility predictability-The role of jumps and regimes

This study mainly focuses on the role of jumps in forecasting Bitcoin volatility using linear and nonlinear mixed data sampling models. The results provide strong evidence that using a forecasting model that incorporates continuous-time jump and two-stage regimes can significantly improve predictive accuracy and achieve high economic gains. Interestingly, the superior forecasting ability of the model with a continuous-time jump is reflected in highly volatile periods, especially in the period of a Black Swan event.

Automated summary

This study mainly focuses on the role of jumps in forecasting Bitcoin volatility using linear and nonlinear mixed data sampling models. The results provide strong evidence that using a forecasting model that incorporates continuous-time jump and two-stage regimes can significantly improve predictive accuracy and achieve high economic gains, particularly during highly volatile periods such as Black Swan events.