Abstract
Robust Orlicz premia have been introduced by Bellini, Laeven and Rosazza Gianin (2018) mainly to take into account in Orlicz-norm based return risk measurement ambiguity with respect to the probabilistic model P, by means of ambiguity averse preferences. In particular, we axiomatized and introduced robustified versions of Orlicz premia and their optimized translation invariant extensions (Haezendonck-Goovaerts risk measures).
In this talk we extend to a dynamic setting the robust Orlicz premia and Haezendonck-Goovaerts risk measures and extensively analyze the properties of the resulting dynamic risk measures. Furthermore, we characterize dynamic Orlicz premia that are time-consistent and prove that the only time-consistent dynamic non-robust Orlicz premia are p-norms. We also show that time-consistency of robust Orlicz premia and of Haezendonck-Goovaerts risk measures are strongly related.
Based on joint work with Fabio Bellini and Roger Laeven.