The purpose of this paper is to apply a belief rule-based (BRB) system to solve the multiasset class portfolio optimisation problems. The BRB system, was developed on the basis of the concept of belief structures and the evidential reasoning (ER) approach, is a generic non-linear modelling and inference scheme. In this paper, the procedures of implementing the BRB system with RiskMetrics WealthBench to portfolio optimisation are discussed in details. Two different ways are proposed to locate the optimal portfolios under constraints supplied by the investors. Numerical studies demonstrate the effectiveness and efficiency of the proposed methodology.
belief rule base evidential reasoning asset class portfolio optimisation