A Practical Guide to Robust Portfolio Optimization

Robust optimization considers uncertainty in inputs to address the shortcomings of mean-variance optimization. We investigate the mechanisms by which robust optimization achieves its goal and give practical guidance regarding its parametrization. We show that quadratic uncertainty sets are preferred to box uncertainty sets, that a diagonal uncertainty matrix with only variances should be used, and that the level of uncertainty can be chosen based on Sharpe ratios. We use examples with the proposed parametrization to show that robust optimization efficiently overcomes the weaknesses of mean-variance optimisation and can be applied in real investment problems like multi-asset portfolio management or robo-advising