基于蒙特卡洛算法的状态依赖期权定价问题研究本科生毕业论文.doc

PAGE 摘 要 当今，世界经济瞬息万变，经济风险随处可见，围绕套期保值的期货、期权及其它金融衍生产品，已经成为经济发展必不可少的工具。特别是，状态依赖型期权由于其自身特点，常常被投资者用来规避风险，但同时需要采用更加合理的数学模型和方法来对期权定价。 本文基于蒙特卡洛算法研究一种状态依赖型期权定价问题，将状态依赖性很强的回望期权作为研究主体，主要研究股票为标的资产的欧式标准回望期权的定价问题。复杂的期权定价模型使得在许多情况下都无法得到期权价值的解析解，而蒙特卡洛方法由于其自身的优点，能够有效的得到期权定价的数值解。 本文首先详细介绍了期权的相关概念和分类、状态依赖期权的概念和分类、维纳过程和蒙特卡洛方法等基础知识，重点介绍了蒙特卡洛算法在期权定价模型中的应用。其次，分别在Black-Scholes模型、CEV模型、Vasicek模型下给出对应回望期权的蒙特卡洛算法，并用matlab软件得到相对应期权的数值模拟，并且在Black-Scholes模型下还进行了理想模拟次数的探讨，在保证时间间隔一定的情况下，得出10000次模拟可以看作是理想的模拟次数。 本文在相应的模型下给出回望期权定价的蒙特卡洛算法，不仅丰富了回望期权定价的研究成果，而且能够在一定程度上对实践具有指导意义。本文最后指出了今后研究需要进一步改进和探讨的方面。 关键词：蒙特卡洛算法； 状态依赖期权； 回望期权； 定价 ABSTRACT Nowadays, the world economy is changing rapidly and economic risks are being everywhere. Many people choose hedging with futures, options and other financial derivatives which have become indispensable tools for the development of economy. Some investors are always using state-dependent options to avoid risks because of its own characteristics. We need to adopt reasonable mathematical models and methods to price options. In this paper, we price the state-dependent options based on Monte Carlo algorithm and choose lookback options as the subject of research. In many categories of the lookback options, we study the pricing of European lookback options with the stock as the underlying asset. Because the option models are more and more complex, in many cases, we cannot get the analytic solution of the options. However, Monte-Carlo method can obtain the numerical solutions of the option pricing with its own advantages. The paper firstly introduces the concepts of options and state-dependent options, Wiener process and Monte Carlo methods, and Monte Carlo algorithm are applied to price options. We respectively give the Monte Carlo algorithm on lookback options under the Black-Scholes model, CEV model, Vasicek model, and also implement algorithm by matlab, and finally run the numerical results. Under the Black-Scholes model, we discuss the idea about the ideal simulatio