中文摘要：

提出了新的基于Black—Scholes模型的混合小波神经网络。隐含波动率是指在市场中观察的期权价格所蕴涵的波动率。基于不同种类的期权价格对波动率的敏感度不同，建立了混合小波神经网络和遗传算法相结合的模型，将期权按钱性进行分类，提出了加权的隐含波动率作为神经网络的输入变量，通过遗传算法来求取不同种类期权的隐含波动率的最优权重。在香港衍生品市场的实证中表明，所提出的模型优于传统的Black～Scholes模型。

英文摘要：

A new hybrid wavelet neural network based on the Black-Scholes model is proposed. Implied volatility is the volatility implied by an option price observed in the market. The sensitivity of the volatility among varied kinds of option price are different. In this work, we build hybrid forecasting models combining hybrid wavelet neural network with genetic algorithm. In using these models, option partition according to moneyness is applied and weighted implied volatility measures are regarded as input of the neural network. The genetic algorithm is used to determine the optimal weight of the implied volatility among different kinds of option. Case study on Hong Kong derivative market shows that these hybrid models are better than the conventional Black-Scholes model.