美式障碍期权定价的总体最小二乘拟蒙特卡罗模拟方法  

Pricing American barrier options by Total Least Squares Quasi-Monte Carlo simulation

作　　者：张利花[1] 张卫国[1] 许文坤[1] 

机构地区：[1]华南理工大学工商管理学院,广东广州510640

出　　处：《数理统计与管理》2013年第5期923-930,共8页Journal of Applied Statistics and Management

基　　金：杰出青年基金项目(70825005);教育部人文社会科学研究规划基金资助项目(07JA630048)

摘　　要：障碍期权的价格依赖于其标的资产的价格路径,实际市场中标的资产的价格变化存在跳跃现象。本文在跳跃扩散模型下使用总体最小二乘拟蒙特卡罗方法(TLSFM)对美式障碍期权定价问题进行了研究。TLSFM使用随机化的Faure序列并结合总体最小二乘回归方法,改进了Longstaff等提出的最小二乘蒙特卡罗模拟方法(LSM)。通过基于TLSFM与LSM和改进的三叉树方法的美式障碍期权定价结果的比较分析,说明了基于TLSFM的美式障碍期权定价具有结果稳定,时效性更强的优势。The price of American barrier options depends on the price path of the underlying asset, in the real financial market there are brusque variations in the underlying assets＇ price. This paper applies Total Least Squares Quasi-Monte Carlo simulation （TLSFM） to price the American barrier options, TLSFM applies stochastic Faure sequences and the Total Least Squares to modify the simple least- squares simulation approach presented by Longstaff. By comparing the results got by TLSFM, LSM and modified Trinomial Trees, respectively, we verify that applying TLSFM to price American barrier options is reasonable, and has two advantages： its results are stable and it is timesaving.

关 键 词：障碍期权 Faure序列 最小二乘估计 跳跃扩散 

分 类 号：O212[理学—概率论与数理统计]