A distributionally robust approach for the risk-parity portfolio selection problem

Risk-parity is one of the most recent and interesting strategies in the portfolio selection area. Considering the mean-standard-deviation risk measure, this paper studies the risk-parity problem under the uncertainty of the covariancematrix. Assuming that the uncertainty is represented by a finite set of scenarios, the problem is formulated as a scenario-based stochastic programming model. Then, since the occurrence probabilities of scenarios are not known with certainty, two ambiguity sets of distributions are considered, and corresponding to each one, a distributionally robust optimization model is presented. Computational experiments on real-world instances taken from the literature confirm the importance of the proposed models in terms of stability, volatility and Sharpe-ratio.