Option Pricing Model Based on Newton-Raphson Iteration and RBF Neural Network Using Implied Volatility

Abstract

As option is a kind of significant financial derivatives, option pricing will affect both the risk and profit of the investment. This paper proposed an option pricing model based on RBF neural network combined with the Newton-Raphson iteration method which is used to obtain the implied volatility. First, considering implied volatility includes investors’ expectation about the changes of future price options. Newton-Raphson iteration method is used to obtain the implied volatility by rolling estimation which is also added into the RBF neural network model. Then, RBF neural network is trained based on Black-Scholes model. Self-organizing learning and the least square method are used to optimize the parameters of RBF neural network. At last, empirical study and analysis with 10 50ETF stock options chosen from Shanghai Stock Exchange market have been performed, the result shows that the accuracy of the proposed model is better than the traditional BP neural network and B-S model and the effect of option pricing using by implied volatility is also better than others.