Robust and Reliable Portfolio Optimization Formulation of a Chance Constrained Problem

Abstract

Abstract We solve a linear chance constrained portfolio optimization problem using Robust Optimization (RO) method wherein financial script/asset loss return distributions are considered as extreme valued. The objective function is a convex combination of portfolio’s CVaR and expected value of loss return, subject to a set of randomly perturbed chance constraints with specified probability values. The robust deterministic counterpart of the model takes the form of Second Order Cone Programming (SOCP) problem. Results from extensive simulation runs show the efficacy of our proposed models, as it helps the investor to (i) utilize extensive simulation studies to draw insights into the effect of randomness in portfolio decision making process, (ii) incorporate different risk appetite scenarios to find the optimal solutions for the financial portfolio allocation problem and (iii) compare the risk and return profiles of the investments made in both deterministic as well as in uncertain and highly volatile financial markets.