波动率预测模型与比较

发布时间：2018-01-11 21:43

本文关键词：波动率预测模型与比较　出处：《上海交通大学》2013年硕士论文　论文类型：学位论文

更多相关文章： 波动率 预测 GARCH模型 隐含波动率

【摘要】：本文选择香港市场恒生指数作为分析标的，旨在比较三种常见的波动率预测模型,GARCH簇模型，BS隐含波动率以及无模型隐含波动率,关于不同长度的时间范围内的波动率的预测能力。隐含波动率与GARCH簇模型相比，，对较长时间范围内的波动率表现出更强的预测精度，尤其是BS隐含波动率，随着预测期限的延展甚至比GARCH模型预测能力更好。由于隐含波动率是基于投资者的市场预期得到，这反映了投资者对较长时间期限内的波动率预测能力较强。另外，无模型隐含波动率的预测精度与市场交易的集中度相关，如果市场集中在平价期权上进行交易，实值和虚值期权会因交易量不足而定价效率不高，以至于无模型隐含波动率的预测能力不强。除此之外，各模型的信息含量也被加以考察，结果表明三者表现出来的信息互有交叠，但是并没有完全覆盖。与前人研究结果的对比表明，香港市场平价期权的定价效率较高，但实值和虚值期权的定价效率取决于市场交易集中度。
[Abstract]:In this paper, the Hang Seng Index of Hong Kong market is chosen as the object of analysis. The purpose of this paper is to compare the implied volatility of the GARCH cluster model with the implied volatility of the GARCH cluster model and the implicit volatility without the model. The ability to predict volatility over time ranges of different lengths. Compared with the GARCH cluster model, the implied volatility shows a stronger prediction accuracy for a long time range, especially for the BS implicit volatility. With the extension of the forecast period, the forecasting ability is even better than the GARCH model, because the implied volatility is based on investors' market expectations. This reflects the strong ability of investors to predict volatility over a longer period of time. In addition, the prediction accuracy of unmodeled implied volatility is related to the concentration of market transactions. If the market focuses on parity options, the real value option and virtual value option will be priced inefficiently because of insufficient trading volume, so that the ability to predict the implied volatility without model is not strong. The information content of each model is also investigated. The results show that the information presented by the three models overlaps with each other, but it is not completely covered. The comparison with previous studies shows that the pricing efficiency of parity options in Hong Kong is high, but the pricing efficiency of real and virtual options depends on the concentration of market transactions.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】