A knowledge based system for evaluation of option pricing algorithms

Presents the design and prototype implementation of a system built around the FINANZIA system that aims in the automated analysis and classification of option pricing algorithms based on experimental data. The main objective is to assist in the generation, storage and evaluation of large amounts of experimental option pricing data and to facilitate the identification of performance properties of the pricing algorithms with respect to the various problems. The analysis of the data is achieved using statistical and inductive logic techniques and the identified properties are used to expand the knowledge base. We demonstrate the use of the system in the context of a case study covering the pricing of American vanilla options in a Black & Scholes (1973) modeling framework.