基于投影降维技术的期权组合非线性VaR模型

发布时间：2018-01-02 14:22

本文关键词：基于投影降维技术的期权组合非线性VaR模型　出处：《管理科学学报》2012年03期 　论文类型：期刊论文

【摘要】：高维期权组合VaR值的计算时间和计算工作量随着市场风险因子维数的增加而迅速增加.为此,引入投影降维技术,用少数几个风险因子来解释高维期权组合总的风险,并结合快速卷积方法,建立了基于投影降维技术的市场风险因子呈厚尾分布情形下的期权组合非线性VaR模型,达到减少计算时间和计算工作量的目的,同时期权组合价值变化的信息又没有太大的损失.数值结果表明,投影降维技术能够达到与快速卷积方法、Monte-Carlo方法差不多的估算精度,而计算效率明显优于快速卷积方法、Monte-Carlo方法,计算时间和计算工作量明显减少.
[Abstract]:The calculation time and workload of VaR value of high-dimensional option portfolio increase rapidly with the increase of the dimension of market risk factor. Therefore, the projection dimension reduction technique is introduced. A few risk factors are used to explain the total risk of high-dimensional option portfolio, and the fast convolution method is used. In this paper, a nonlinear VaR model of option combination based on projection dimensionality reduction technique is established, in which the market risk factor is distributed with a thick tail, which can reduce the calculation time and workload. At the same time, the information of option portfolio value change is not much lost. The numerical results show that the projection dimensionality reduction technique can achieve the same accuracy as the fast convolution method and Monte-Carlo method. The computational efficiency is obviously better than that of the fast convolution method and Monte-Carlo method, and the computational time and workload are obviously reduced.
【作者单位】：浙江大学管理学院;浙江财经学院金融学院;
【基金】：国家自然科学基金资助项目(70771099;71171176) 浙江省哲学社会科学重点研究基地社科规划课题重点资助项目(10JDGZ02Z)
【分类号】：F224;F830.9
【正文快照】： 0引言当市场风险因子数目很大时,度量期权组合风险的计算时间和计算工作量随着市场风险因子维数的增加而迅速增加,使得建模和数值计算十分的困难(Wiberg[1]).这时人们自然想到简化或减少风险因子的维数而对期权组合价值变化的内容又没有太大的损失,通过度量低维的与其相似的

【参考文献】