当前位置:主页 > 管理论文 > 信贷论文 >

路径依赖型期权定价模型和方法研究

发布时间:2018-04-18 05:00

  本文选题:路径依赖型期权 + 跳跃扩散过程 ; 参考:《华南理工大学》2013年博士论文


【摘要】:期权作为一种重要的衍生产品,在金融市场上起着重要的作用,作为套期保值的一种重要工具,能很好的规避风险,指导市场参与者进行投资。随着我国金融市场的不断发展完善,以及对金融市场风险的控制的需求,期权在中国市场得到了进一步发展,中国的期权市场也在不断的完善中。路径依赖型期权应市场的需求而产生,由于它的一些特性,满足了市场参与者的要求而成为期权市场交易的主要产品。如何建立合理的期权定价模型,并对路径依赖型期权进行合理的定价已经成为金融机构和监管部门关注的热点。自上个世纪70年代以来,众多学者研究了几何布朗运动模式下期权定价及其定价方法问题,对路径依赖型期权,尤其是美式期权,在几何布朗运动下的定价方法做了大量的研究。然而近年来,对资产收益率的实证研究表明金融资产的对数收益率的分布并不是对称的,是有偏的,并且呈现尖峰厚尾性,波动率也不是一个常数,而是呈微笑型,并且存在波动率聚类现象。这些性质说明传统的布朗运动已不能很好的刻画金融资产的随机现象,而缓增稳定分布的不对称性,以及厚尾性和GARCH模型对波动率微笑型的描述恰好能够刻画金融资产的这些随机现象。因此在缓增稳定分布GARCH模型下对期权进行定价并在模型下对路径依赖型期权的定价方法进行研究,能够有效指导金融市场投资者,对确定公平合理的期权价格,具有一定的理论和现实意义。 由于金融市场的波动性,以及市场参与者具有主观性,不同参与者对市场参数具有不同的看法,,所以在对金融市场的资产收益进行建模时,模型中的参数假设为清晰数与市场的实际情况不符。为了更好地刻画市场的特性,并且兼顾资产收益分布的上述特性,本文对资产收益的建模进行了研究,建立了模糊环境下的期权定价模型,采用缓增稳定分布为资产的收益建模。并在模型的基础上,对路径依赖型期权的定价方法进行了研究。主要工作和创新归纳如下: 第一,在跳跃扩散模型下,研究了障碍期权,回望期权的定价问题,并对美式障碍期权,美式回望期权的定价方法进行了研究。 已有的对期权定价问题的研究都集中在几何布朗运动下的定价模型研究。但是大量的实证研究表明,资产收益的分布是有偏的,并且资产的价格变化存在跳跃现象,不是连续变化的。首先,利用跳跃扩散模型描述资产的价格变化过程。其次,障碍期权和回望期权的价格都与标的资产的价格路径有关,从模型中把期权的定价公式的解析解求出来比较困难,这里我们对已有定价方法进行改进,提出了总体最小二乘拟蒙特卡洛模拟方法对期权定价。数值分析表明,改进后的期权定价方法时效性更强,运算精度更高。 第二,结合实际市场资产收益分布的特性,我们应用缓增稳定分布对资产的价格过程建模,缓增稳定过程能很好的刻画资产收益的有偏性,尖峰厚尾现象。同时,缓增稳定分布与GARCH模型结合建立资产收益的动态过程,GARCH模型解决了波动率的微笑型和聚类现象。所以二者的结合能更好的刻画资产的价格过程。 GARCH模型能很好的解决波动率聚类现象,但是实际金融市场的分布是有偏的,具有尖峰厚尾性,用正态分布刻画与实际市场不符。本文把GARCH模型的残差分布由正态分布改为缓增稳定分布,建立了缓增稳定分布GARCH模型,这类模型能最大程度的拟合市场资产价格过程。K-S检验和A-D检验均表明,缓增稳定分布GARCH模型与实际市场最接近。 第三,提出马氏链方法对缓增稳定分布GARCH模型的美式期权定价,并且对方法本身的收敛性进行证明。马氏链方法比传统的最小二乘蒙特卡洛模拟定价方法优越:时效性更强。 由于本文采用缓增稳定分布和GARCH刻画期权标的资产价格变化的行为模式,如何缓增稳定分布GARCH模型下的期权,尤其是美式期权进行定价是关键问题。传统的期权定价方法是蒙特卡洛模拟和二叉树计算。本文采用马氏链计算方法,通过转移概率矩阵把期权定价公式计算出来,不仅能为欧式期权定价,而且解决了美式期权定价问题。马氏链方法与最小二乘蒙特卡洛相比,运行时间较短,提高了运算速度。 第四,在模糊环境下对期权定价问题进行研究,建立了模糊双指数跳跃扩散模型,应用清晰可能性均值对模糊双指数跳跃扩散期权定价模型进行清晰化,并给出了模糊双指数跳跃扩散模型的求解方法。 期权定价的已有研究都是在假设标的资产价格行为模式服从几何布朗运动下进行的,在清晰数环境下。本文通过采用双指数跳跃扩散模型来刻画期权标的资产价格变化的行为模式,并且在模糊数的环境下,对市场的一些参数进行刻画,建立了模糊双指数跳跃扩散期权定价模型,进一步结合模糊数的清晰可能性均值,对模型进行清晰化,建立了清晰可能性均值下的双指数跳跃扩散定价模型。应用研究表明:模糊环境下的双指数跳跃扩散期权定价模型更符合实际的金融市场,定价的结果比清晰环境下的期权定价模型提供了更多的有用信息,供投资者参考。因此模糊环境下的期权定价模型的引入完善了期权产品的定价。
[Abstract]:As an important derivative , the option plays an important role in the financial market . As an important tool for hedging , it can avoid the risk and guide the market participants to invest .

Because of the volatility of financial markets and the subjectivity of market participants , different players have different views on market parameters , so the parameters in the model are assumed to be consistent with the actual situation of the market . In order to better characterize the market characteristics , and to balance the above - mentioned characteristics of the distribution of asset returns , this paper studies the modeling of asset returns , establishes an option pricing model under the fuzzy environment , and studies the pricing methods of the path - dependent options on the basis of the model .

First , under the model of jump diffusion , the pricing problem of barrier options and return options is studied , and the pricing methods of American barrier options and American return options are studied .

The research on options pricing has focused on pricing model research under geometric Brownian motion . However , a large number of empirical studies show that the distribution of asset returns is biased , and the price change of assets is not continuous . First , the pricing formula of the options is compared with the price path of the target asset . Secondly , the pricing formula of the options is improved by using the jump diffusion model . The numerical analysis shows that the improved option pricing method is more efficient and the operation precision is higher .

Second , according to the characteristics of the real market asset income distribution , we apply the gradual increase and stability distribution to modeling the price process of the asset , and the gradual increase and stability process can depict the biased and peak thickness tail phenomena of the asset gains . At the same time , the gradual increase and stability distribution and the ARCH model are combined to establish the dynamic process of the asset returns , and the ARCH model solves the smile type and clustering phenomenon of the volatility . Therefore , the combination of the two can better characterize the price process of the asset .

In this paper , the distribution of the real financial markets is biased , and the distribution of the real financial markets is biased . The distribution of the residual distribution of the real market is inconsistent with the real market . In this paper , the distribution of the residual distribution of the ARCH model is changed from the normal distribution to the slow - increasing stable distribution , and the gradual - increasing and steady - distribution model is established . The K - S test and the A - D test show that the slow - increasing stability distribution is the closest to the actual market .

Thirdly , this paper puts forward the American option pricing of the Markov chain method to the slow - increasing and stable distributions , and proves the convergence of the method itself . The Markov chain method is superior to the traditional least square Monte Carlo simulation pricing method : the timeliness is stronger .

The paper uses Markov chain method to calculate the option pricing formula by transfer probability matrix , which not only can set the European option pricing , but also solve American option pricing problem .

Fourthly , under the fuzzy environment , the paper studies the option pricing problem , establishes a fuzzy double exponential jump diffusion model , and makes a definition of the fuzzy double exponential jump diffusion option pricing model by means of the clear possibility mean value , and gives the solution method of the fuzzy double exponential jump diffusion model .

The existing research on option pricing is carried out under the assumption that the target asset price behavior pattern is subject to the geometric Brownian motion , and under the clear number environment , the paper describes the behavior pattern of the asset price change of the option subject by using the double exponential jump diffusion model , and establishes a double exponential jump diffusion pricing model under the fuzzy number .

【学位授予单位】:华南理工大学
【学位级别】:博士
【学位授予年份】:2013
【分类号】:F224;F830.9

【引证文献】

相关硕士学位论文 前1条

1 李淑琦;基于Heston随机波动模型的近似精确仿真技术研究[D];广东工业大学;2014年



本文编号:1766881

资料下载
论文发表

本文链接:https://www.wllwen.com/guanlilunwen/bankxd/1766881.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户2d574***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com