In recent years, cryptocurrencies have gained investor attention for their extreme volatility, but this has introduced financial risks that require accurate prediction models. Therefore, we propose the SHARV-MGJR model, which incorporates both ‘good-bad’ volatility, leverage effects, and current return information to enhance the accuracy of cryptocurrency market volatility predictions. Empirical results demonstrate that compared to GARCH-type models, the SHARV-MGJR model exhibits superior predictive accuracy in forecasting cryptocurrency market volatility. Furthermore, robustness tests confirm the superiority of the SHARV-MGJR model in predicting cryptocurrency market volatility.

中文翻译：

---

“好坏”波动性和杠杆效应能否改善对加密货币波动性的预测？——来自 SHARV-MGJR 模型的证据


近年来，加密货币因其极端的波动性而受到投资者的关注，但这也带来了需要准确预测模型的金融风险。因此，我们提出SHARV-MGJR模型，该模型结合了“好坏”波动性、杠杆效应和当前回报信息，以提高加密货币市场波动性预测的准确性。实证结果表明，与 GARCH 类型模型相比，SHARV-MGJR 模型在预测加密货币市场波动方面表现出优异的预测准确性。此外，稳健性测试证实了 SHARV-MGJR 模型在预测加密货币市场波动方面的优越性。