论文部分内容阅读
结合日内跳跃识别方法和马尔可夫机制转换模型,对已实现波动率异质自回归模型(HARRV)进行拓展,以刻画连续波动、跳跃波动以及不同方向跳跃波动对未来波动影响的差异和波动的结构转换特征,并运用该模型对上证综指和深证成指高频数据进行实证分析。研究结果表明:在短期内,连续波动和跳跃波动对未来波动影响具有显著的差异;负向跳跃和正向跳跃往往同时发生且幅度相当,但负向跳跃波动对未来波动的影响更大;在不同波动状态下,历史波动对未来波动的影响存在较为明显的差异。MCS检验结果显示,区分跳跃波动方向和考虑波动的结构转换特征可以显著提升模型的样本内和样本外的预测能力。
Combined with intra-day jump recognition method and Markov mechanism transformation model, the realized volatility heterogeneous autoregressive model (HARRV) is expanded to characterize the differences and fluctuations of continuous fluctuations, jump fluctuations and future fluctuations of different directional jump fluctuations Structure transformation features, and use the model to empirical analysis of the Shanghai Composite Index and the Shenzhen Component Index high-frequency data. The results show that in the short term, continuous fluctuations and jump fluctuations have significant differences on future fluctuations. Negative jumps and forward jumps tend to occur at the same time with similar amplitudes, but negative jump fluctuations have a greater impact on future fluctuations. Under fluctuating conditions, there are obvious differences in the impact of historical fluctuations on future fluctuations. The results of MCS test show that distinguishing the directions of jumps and structural transformations considering fluctuations can significantly improve the predictive ability both within and outside the model.