Robust utility maximization for complete and incomplete markets (Q2488473)

This paper is devoted to the robust utility maximization problem for some convex set of reasonable equivalent local martingale measures. The payoffs are required to be obtainable from dynamic trading in the underlying assets. Dual characterizations are given for general solutions for both a complete and an incomplete market models. A dual characterization is given also for a closely related problem of the minimization of expenditures given a minimum level of robust expected utility. The main tool are the techniques developed for \(f\)-divergences. With the help of this notion the robust problem is reduced to the classical problem of utility maximization under a certain measure: the reverse \(f\)-projection. This minimizing measure has the property that for a complete market the classical utility maximization problem is equivalent to the robust problem.