Bayes推断和神经网络求解美式回望期权的隐含波动率  

Solving Implied Volatility of American Lookback Options byBayesian Inference and Neural Network

作　　者：陶李 朱本喜[2] 钱译缘 徐嘉琪 TAO Li;ZHU Benxi;QIAN Yiyuan;XU Jiaqi(College of International Business,Hainan University,Haikou 570228,China;College of Mathematics,Jilin University,Changchun 130012,China)

机构地区：[1]海南大学国际商学院,海口570228 [2]吉林大学数学学院,长春130012

出　　处：《吉林大学学报（理学版）》2024年第6期1363-1369,共7页Journal of Jilin University:Science Edition

基　　金：国家自然科学基金面上项目(批准号:12271207).

摘　　要：首先,用原始对偶活跃集方法求解期权定价正问题,将相应的数值解作为监督学习的输出,然后用训练好的神经网络替代期权定价正问题模型.其次,结合Bayes推断与神经网络进行Metropolis-Hastings采样,求解隐含波动率反问题.该方法减少了采样过程中正问题计算量庞大的问题,从而加速了反问题求解过程.Firstly,we used the original dual active set method to solve the forward problem of option pricing,with the corresponding numerical solutions as the output for supervised learning,and then replaced the forward problem model of option pricing with a well-trained neural network.Secondly,we combined Bayesian inference with neural networks for Metropolis-Hastings sampling to solve the inverse problem of implied volatility.This method reduced the problem of large computational complexity of the forward problem during the sampling process,thereby accelerating the solution process for the inverse problem.

关 键 词：隐含波动率 BAYES推断 神经网络 替代模型 

分 类 号：O241.8[理学—计算数学]