基于加权最小二乘拟蒙特卡罗的美式期权定价

美式期权定价具有后向迭代搜索特征, 在最小二乘拟蒙特卡罗模拟(least-squares quasi-Monte Carlo, LSM)方法的基础上, 本文通过随机Faure序列以及对偶变数法增加抽样数目, 达到减小模拟方差的目的,用其计算标的资产价格, 然后用加权最小二乘法进行回归, 得到了加权最小二乘拟蒙特卡罗(weighted least-squares quasi-Monte Carlo, WLSQM)方法. 从期权价值、标准差、运行时间几个方面比较方法的优劣, 得出WLSQM 方法比LSM 方法估计效果更优的结果, 验证了WLSQM方法在美式期权定价上的有效性.

Pricing American options with weighted least-squares quasi-Monte Carlo

American options pricing has the backward feature of iterative search.Based on the least- squares Monte Carlo(LSM)method,this paper employs Faure sequences and doubles the sample' s number by using antithetie variate method to decrease the variance of simulation.Then,underlying assets are valued.Thus,a weighted least-squares quasi-Monte Carlo(WLSQM)method is proposed by weighted least-squares regression.Compared the two methods for option value,standard error and computation cost,WLSQM method is better than...