Enhancing portfolio management using artificial intelligence: literature review.

Area of Science:

• Quantitative Finance
• Computational Finance
• Financial Technology (FinTech)

Background:

• Portfolio management aims to balance risk and reward through optimal asset allocation.
• Traditional frameworks like Markowitz's mean-variance theory have guided asset allocation for decades.
• Advances in artificial intelligence offer new capabilities for complex financial problems.

Purpose of the Study:

• To review state-of-the-art artificial intelligence approaches in investment portfolio management.
• To analyze how AI transforms the core steps of portfolio management: planning, execution, and feedback.
• To demonstrate the application of post-hoc explanations for asset allocation in AI-driven finance.

Main Methods:

• Literature review of current AI applications in portfolio management.
• Analysis of AI's impact on the planning, execution, and feedback stages.
• Case study illustrating post-hoc explanations for AI-based asset allocation.

Main Results:

• AI significantly enhances traditional portfolio management processes.
• AI offers advanced solutions for complex asset allocation challenges.
• Post-hoc explanations are crucial for addressing transparency and explainability in AI finance.

Conclusions:

• Artificial intelligence is a transformative force in modern portfolio management.
• Explainable AI (XAI) is essential for regulatory compliance and trust in financial applications.
• The integration of AI, particularly XAI, promises greater transparency in the European investment business.