Qutrit-Based Quantum-Inspired Optimization Model on Real-World Portfolio Optimization

Quantum three-level qutrit systems can achieve high-fidelity operations and are favorable for quantum error correction, which is crucial in the current NISQ era. Before the advent of fault-tolerant computers, quantum-inspired optimization (QIO) shows potential as a practical near-term application of quantum computing to address real-world problems. Consequently, we have developed a novel encoding mechanism leveraging the quantum three-level qutrit concept to propose a qutrit-based QIO for the quantum finance model. This approach comprehensively addresses financial applications by encoding additional strategic information, unlike traditional methods that only consider binary buy-or-not-buy decisions. In a volatile eco-nomic environment, when the corresponding properties of qutrits are simulated, multiple strategies can be employed to stabilize investment efficiency, including both long-selling (LS) and short-selling (SS) strategies. The experimental results demonstrate that the proposed qutrit-based model can construct a portfolio with superior performance compared to the two single strategies when using real-world financial data. Furthermore, this study reveals that the fluctuations from LS and SS strategies can be complementary within a hybrid strategy, contributing to a significant reduction of risk. The proposed qutrit-based method uses the advantages of quantum behavior in a large solution space, searches for potential solutions efficiently in real-world applications, and aids investors in making an informed decision.