An Option Pricing Model That Combines Neural Network Approach and Black Scholes Formula
The Black & Scholes formula for theoretical pricing of options exhibits certain systematic biases, as observed prices in the market differs from the formula. A number of studies attempted to reduce these biases by incorporating a correction mechanism in the input data. Amongst non-parametric approaches used to improve accuracy of the model, Artificial Neural Networks are found as a promising alternative. The study made an attempt to improve accuracy of option price estimation using Artificial Neural Networks where all input parameters are adjusted by suitable multipliers. The values of these multipliers were determined using known data that minimises errors in valuation. The study was carried out using Nifty call option prices quoted on National Stock Exchange for the period 01-Jul 2008 to 30-Jun-11 covering three years.
