Pricing and hedging European options with discrete-time coherent risk (Q2463721)

The author considers pricing and hedging in discrete-time models. He chooses dynamic risks because they allow to use backward induction, and restricts attention to coherent risks since coherent risk measures lead to more explicit results. The major goal of the paper is to obtain as explicit as possible forms of the risk-based upper/lower prices as well as superhedging/subhedging strategies in the discrete-time setting. The fundamental theorem of asset pricing is proved for the risk-based \textit{No Good Deals} condition within the framework of a general infinite-dimensional model. Then the risk-based upper and lower price processes are defined for a stream of cash flows within the framework of a general multidimensional model. Probabilistic and geometric representations are provided for these price processes. For a wide class of models with some Markov property, in particular, for GARCH models, more concrete formulas are given. A sufficient condition for the uniqueness of a hedging strategy is established and numerical algorithm is presented.