基于小波神经网络的期权定价模型

Black-Scholes模型所要求的假设条件在真实的市场条件下往往不能满足.提出了一种新的应用小波神经网络进行预测的欧式期权定价模型.将期权按钱性进行分类, 以一种新的加权的隐含波动率作为神经网络的输入变量,通过小波神经网络模型、BP网络模型和Black-Scholes模型来预测香港恒指买权的价格.实证结果表明,将一种加权的隐含波动率作为输入变量的小波神经网络模型优于Black-Scholes模型和其他神经网络模型.因此该模型可以更有效地预测欧式期权价格.

Option pricing models based on wavelet neural network

The option markets often violate most of the Black-Scholes model's assumptions.A new European option price forecasting model by applying wavelet neural network is proposed.In such an approach,option partition according to moneyness is applied.Herein,the ability is compared among the wavelet neural network models,the back propagation(BP) neural network models and the Black-Scholes model to price HSI(Hang Seng index) call options.An implied volatility measure is used.The experimental results show that when a weighted implied volatility measure is regarded as an input,the performance of the wavelet neural network is better than that of other neural network models and the Black-Scholes model.Therefore,this model can be used to foresee European option price more effectively.