Pruning Pareto optimal solutions for multi-objective portfolio asset management

Budget allocation problems in portfolio management are inherently multi-objective as they entail different types of assets which performance metrics not directly comparable. Existing asset methods that either consolidate multiple goals to form a single objective (a priori) or populate Pareto optimal set posteriori) may be sufficient because decision maker (DM) possess comprehensive knowledge the problem domain. Moreover, current techniques often present with too many options, making it counter-productive. In order provide DM diverse yet compact solution set, this paper proposes three-step approach. first step, we employ approximation functions capture investment-performance relationships at asset-type level. These simplified then used inputs for optimisation model second step. final solutions generated by selected evolutionary algorithm pruned clustering method. To measure spread representative over front, two novel indicators based on average Euclidean distance and cosine similarity between original solutions. Through numerical examples, demonstrate approach can maintain high integrity front. We also put forward suggestions choosing appropriate functions, pruning methods, indicators.