Title:Sensitivity of causal distributionally robust optimization

Abstract:We study the causal distributionally robust optimization (DRO) in both discrete- and continuous- time settings. The framework captures model uncertainty, with potential models penalized in function of their adapted Wasserstein distance to a given reference model. Strength of the penalty is controlled using a real-valued parameter, which is simply the radius of the uncertainty ball in the special case of an indicator penalty. Our main results derive the first-order sensitivity of the value of causal DRO with respect to the penalization parameter, i.e., we compute the sensitivity to model uncertainty. Moreover, we investigate the case where a martingale constraint is imposed on the underlying model, as is the case for pricing measures in mathematical finance. We introduce different scaling regimes, which allow us to obtain the continuous-time sensitivities as nontrivial limits of their discrete-time counterparts. We illustrate our results with examples.
Our proofs rely on novel methods. In particular, we introduce pathwise Malliavin derivative, which agree a.s. with its classical counterpart under the Wiener measure, and we extend the adjoint operator, the Skorokhod integral, to regular martingale integrators and show it satisfies a stochastic Fubini theorem.