Multi-Objective Stochastic Optimal Asset Allocation for DC Pension under Unpredictable Non-Market Disturbances

Abstract

In this paper, we mainly solve the optimal asset allocation problem of DC(defined-contribution) pension under unpredictable non-market disturbances. There are only two types of assets which are allowed to be invested: a risk-free cash bond and a risky stock. Particularly, the pension managers in this paper aim to achieve the minimization of the accumulated deviations between the manager's pre-set target and the actual fund scale with less management cost and make the DC pension plan robust under unpredictable non-market disturbances. To achieve the goals, the definition of robustness performance index with respect to DC pension plan is proposed, and a stochastic multi-objective optimal asset allocation model is established based on the definition. By applying the Itô formula and the matching method, the multi-objective optimal portfolio problem is approximate into a sequence optimization problem with LMI-constrained. To solve the sequence optimization problem, a sequence optimization algorithm is developed in this paper. Finally, we give a numerical simulation to demonstrate the effectiveness of our proposed model.