A multiobjective genetic programming approach for pricing and hedging derivative securities

Genetic programming has become increasingly important in the broad field of finance. Due to the fact that in the context of pricing derivatives a lot of models are not amenable to analytical exact solutions, this is especially true for the research area of contingent claim pricing. Previous work in this field almost certainly always focuses on pricing such derivatives. In doing so, the implied hedging performance is totally ignored due to the lack of an analysis of price sensitivities which are fundamental building-blocks in hedging-strategies. In this contribution we apply a multiobjective genetic programming approach to the American put pricing problem and evaluate our individuals based on the pricing as well as on the hedging performance by means of symbolic differentiation.