中文摘要：

采用修正的已实现门阀多次幂变差研究沪深300股指期货已实现波动率的跳跃特征，并通过构建考虑跳跃的AHAR—CTCJ模型研究沪深300股指期货已实现波动率的跳跃成分对股指期货市场未来波动率预测的影响。结果表明：沪深300股指期货波动率的连续成分存在较强的长记忆性，而离散跳跃序列的长记忆较弱，但仍具有一定的可预测性；跳跃的久期序列的自相关性较强，而规模序列的自相关性不显著；显著的离散跳跃对沪深300股指期货日、周以及月的已实现波动率的预测都存在显著的正向影响，AHAR—CTCJ模型能显著提高股指期货市场波动率的预测精度，尤其是标准差形式和对数形式的AHAR—CTCJ模型。

英文摘要：

We utilize corrected realized threshold multipower variation to investigate the jump dynamics of the realized volatility of CSI 300 stock index futures, and AHAR-C _ TCJ model with jump to analyze the impact of jumps on realized volatility forecast for CSI 300 stock index futures. We find that the continuous composition of realized volatility of CSI 300 stock index futures have strongly long memory, the discretely jumps have little long memory, but a certain degree of predictability was still found in jumps, the correlation of jump duration is strong, but the correlation of jump size is not significantly. In addition, the significant jumps have positive and significant impacts on daily, weekly and monthly realized volatility forecast of CSI 300 Stock Index Futures, and the AHAR-C _ TCJ model provides a significantly superior forecasting ability for the volatility of Stock Index Futures, especially the square-root and logarithmic model.