Robo-Advising in Motion: A Model Predictive Control Approach

2026-01-14

AI Summary

This study addresses the limitations of traditional robo-advisory systems, which typically rely on static single-period asset allocation and struggle to adapt to dynamic markets while accounting for practical constraints such as transaction costs and turnover. To overcome these challenges, the paper introduces model predictive control (MPC) into robo-advisory for the first time, proposing a multi-period dynamic asset allocation framework. The approach integrates a hidden Markov model with the Black-Litterman methodology to forecast returns and covariances, while explicitly incorporating transaction costs and turnover constraints into the optimization process to achieve dynamic mean-variance and dynamic risk-budgeting objectives. Numerical experiments demonstrate that the proposed MPC-based strategy significantly outperforms myopic alternatives: it yields flexibly diversified portfolios under the mean-variance formulation and produces parameter-robust, smoothly evolving allocations under risk budgeting, effectively balancing practical feasibility with dynamic adaptability.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Mixed Discrete/Continuous Search
• Multiagent Systems: Mechanism Design

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments

Abstract

Robo-advisors (RAs) are automated portfolio management systems that complement traditional financial advisors by offering lower fees and smaller initial investment requirements. While most existing RAs rely on static, one-period allocation methods, we propose a dynamic, multi-period asset-allocation framework that leverages Model Predictive Control (MPC) to generate suboptimal but practically effective strategies. Our approach combines a Hidden Markov Model with Black-Litterman (BL) methodology to forecast asset returns and covariances, and incorporates practically important constraints, including turnover limits, transaction costs, and target portfolio allocations. We study two predominant optimality criteria in wealth management: dynamic mean-variance (MV) and dynamic risk-budgeting (MRB). Numerical experiments demonstrate that MPC-based strategies consistently outperform myopic approaches, with MV providing flexible and diversified portfolios, while MRB delivers smoother allocations less sensitive to key parameters. These findings highlight the trade-offs between adaptability and stability in practical robo-advising design.

Problem

• robo-advising
• dynamic asset allocation
• multi-period optimization
• transaction costs
• portfolio management

Innovation

• Model Predictive Control
• Robo-Advising
• Dynamic Asset Allocation
• Black-Litterman
• Hidden Markov Model