ECCO XXXVI CONFERENCE MAY 2023

“The Structure of Mean-Variance and Max Entropy Frontiers In Portfolio Optimization” authored by professors Yiannis Dimoticalis and Takis Varelas was presented by Yiannis in the6^th Conference of the European Chapter on Combinatorial Optimization be held in Chania, Crete, Greece, from Thursday 11^th to Sunday 14^th May 2023. ECCO (European Chapter on Combinatorial Optimization) is a Working Group of EURO (Association of European Operational Research Societies) that provides an excellent opportunity to discuss recent and important issues in Combinatorial Optimization and its applications.

The traditional Markowitz Mean–Variance method for optimizing a portfolio is based on risk (variance) minimization. The combinations of portfolio assets lead to the associated Mean-Variance frontier. The Max Entropy approach is based on maximizing entropy as a way to measure portfolio differentiation. Geometrically the structure of the Max Entropy frontier is similar to that of the Mean-Variation frontier. Both frontiers are the answer to a specific optimization problem with certain constrains. The cases of 2, 5, and 10 assets portfolios are used to show the similarities and differences of the two frontiers. The implications for portfolio managers are discussed using a new index for the Max Entropy frontier that is comparable to the Sharpe index of the Mean-Variance method.