Revising Portfolio with Preferences via Higher-Order Group Multi-Role Assignment

Portfolio creation and management are fundamental and important investment services. An investment manager is often responsible for dealing with many clients. Thus, s/he needs tools appropriate to creating, managing and matching portfolios with clients, according to market conditions. In planning, one financial product can be assigned to many, but different clients and one client may buy many, but different products. It is in fact a higherorder investment problem that involves many to many (M2M) assignment in the process. The manager must establish a mix of products, for various levels of return, that suit a client's risk tolerance. This paper solves the higherorder M2M assignment problem via the higher-order group multi-role assignment (HO-GMRA). Based on the concise formalization of Role-Based Collaboration (RBC) and its E-CARGO model, a practicable multi-object optimization approach is proposed, and its kernel is an x-ILP (extended Integer Linear Programming) planning method. Method verification is achieved by simulation experiments with respect to a real-world problem. The experimental results demonstrate the practicability of the proposed solutions.