Neural networks models for the prediction of stock return volatility

In this paper we will focus on the use of artificial neural networks (ANNs) for the prediction of stock return volatilities. This stock return volatility is an important and unknown parameter in the Black and Scholes formulas (1973) for the calculation of the value of stock options. We will demonstrate that it is feasible to implement a NARMAX-like structure with neural networks, and to use this architecture for predicting the stock return volatility.