A Robust Risk-based Tactical Asset Allocation

In this paper we define and compare different versions of robust, in the sense of Robust Optimization, and non robust portfolio selection models alternatively based on the use of different risk measures. This with the aim to take account of investors' asymmetric preferences in profits and losses together with the goal of having solutions less dependent on the parameter uncertainty. The empirical implementation considers the time series of the monthly prices of some representatives benchmarks in a time period characterized by a very particular set of financial events and therefore an ideal time to test the different portfolios strategies related to the alternative models. We show that the robust CVaR approach is preferable compared with the others and with the risk-free portfolio. The results can have very interesting applications in the field of the asset management industry.