Computational intelligence approach to capturing the implied volatility

Abstract

In this paper, a Computational Intelligence Approach and more particularly a neural network is used to learn from data on the Black-Scholes implied volatility. The implied volatility forecasts, generated from the Neural Net, are converted to option price using the Black-Scholes formula. The neural network option pricing capabilities are shown to be superior to the Black-Scholes and the GARCH option-pricing model. The neural network has also shown that it is able to reproduce the implied volatility well into the future whereas the GARCH option-pricing model shows deterioration in the implied volatility with time.