MULTIOBJECTIVE MODEL PREDICTIVE CONTROL IN STOCK PORTFOLIO OPTIMIZATION

This paper proposes a Multi-Objective Model Predictive Control (MO-MPC) framework for stock portfolio optimization, designed to achieve an optimal balance between return maximization and risk minimization in volatile financial markets. This approach integrates Stochastic Model Predictive Control (SMPC) to predict asset returns and dynamically adjust portfolio allocation based on a discrete-time state-space model. The optimization problem is formulated as a multi-objective optimization and is solved using Multi-Objective Particle Swarm Optimization (MOPSO). Simulation results show that the MO-MPC approach significantly outperforms conventional methods regarding wealth maximization and risk minimization. Moreover, SMPC performs better than MOPSO in maximizing portfolio value and reducing risk. These findings confirm the potential of SMPC as an adaptive and reliable strategy for financial decision-making under uncertainty.