ASSESSMENT OF NUMERICAL METHODS FOR DERIVATIVE PRICING: REVISION AND APPLICATION TO THE LEASE PURCHASE OF TELEBRÁS PN

Abstract

The goal of this work is twofold: (i) to review numerical methods to price derivatives; (ii) to compare numerical methods assuming that market prices are reflected in the Black and Scholes formula. We apply these models to price call option on Telebrás shares. Accuracy and computational costs were used to compare the following methods: binomial, Monte Carlo, and finite difference. Our results indicate the good accuracy at low cost for binomial methods, followed by Monte Carlo and finite difference methods. However, Monte Carlo could be used when the derivative depends on two or more underlying assets. In addition, finite difference method should be used when the solution for the partial differential equation whose solution is the derivative price.