A robust hedging algorithm

Abstract Hedging the risk of writing options is a major concern in finance considering the potentially huge losses in derivatives trading. We present two hedging strategies which determine the optimum number of shares that minimizes the worst-case potential hedging error under transaction costs, given some predefined sources of uncertainty. We present a minimax algorithm as a computational tool for these strategies and discuss its performance from the point of view of optimization. It is shown that, depending on the hedging strategy, the performance of the algorithm is determined by the Hessian of the objective function, which in turn is determined by the option pricing model embedded within the objective function.