论文部分内容阅读
本文利用上证综指在2005～2009年内的日间高频数据,通过已实现波动率这一概念对我国股市在这5年间的波动特性做了研究。进一步地,根据已实现波动率序列的统计特征,对其进行长记忆建模,并对模型的波动率预测效果与常规GARCH模型的预测效果做了对比分析。基于上证综指的研究结果表明,利用了日间高频信息的波动率模型在波动率预测上,比仅利用了收盘信息的GARCH模型更有优势。
This paper uses the daytime high frequency data of Shanghai Composite Index from 2005 to 2009 and studies the volatility characteristics of China’s stock market in these 5 years through the concept of realized volatility. Further, according to the statistical characteristics of the realized volatility series, the long-memory model is constructed and the forecasting effect of the model’s volatility is compared with that of the conventional GARCH model. Based on the results of the Shanghai Composite Index, the volatility model using daytime high frequency information is more advantageous than the GARCH model that uses only closing information for volatility forecasting.