Dynamic Investment Portfolio Optimization under Constraints in the Financial Market with Regime Switching using Model Predictive Control

TL;DR

This paper develops a model predictive control approach for dynamic portfolio optimization under trading constraints, considering regime-switching market dynamics modeled by Markov-modulated processes, and tests it on real financial data from diverse markets.

Contribution

It introduces a novel MPC-based method for portfolio optimization under constraints with regime-switching dynamics, validated on multiple real-world financial markets.

Findings

Effective feedback trading strategies derived from MPC.

Successful application to diverse market data sets.

Demonstrated robustness across different economic regimes.

Abstract

In this work, we consider the optimal portfolio selection problem under hard constraints on trading volume amounts when the dynamics of the risky asset returns are governed by a discrete-time approximation of the Markov-modulated geometric Brownian motion. The states of Markov chain are interpreted as the states of an economy. The problem is stated as a dynamic tracking problem of a reference portfolio with desired return. We propose to use the model predictive control (MPC) methodology in order to obtain feedback trading strategies. Our approach is tested on a set of a real data from the radically different financial markets: the Russian Stock Exchange MICEX, the New York Stock Exchange and the Foreign Exchange Market (FOREX).