Problem
Research questions and friction points this paper is trying to address.
Agentic AI
Strategic Asset Allocation
Institutional Asset Management
Autonomous Agents
Investment Policy Statement
The Self Driving Portfolio: Agentic Architecture for Institutional Asset Management
🤖 AI Summary
This work proposes an autonomous asset allocation system comprising approximately 50 specialized agents, addressing the limitations of traditional human-driven approaches that struggle to efficiently integrate diverse models and dynamic market feedback. For the first time, the system adopts the Investment Policy Statement as a unified governance framework, enabling seamless collaboration between human oversight and AI-driven decision-making. It incorporates capital market assumption generation, over twenty portfolio construction methodologies, and an inter-agent peer-review voting mechanism. A dedicated research agent introduces novel strategies, while a meta-agent autonomously refines code and prompts based on historical performance. Through a multi-agent architecture, self-directed code rewriting, and backtesting validation, the system continuously evolves, demonstrating superior robustness and adaptability in both forecasting and allocation compared to single-model approaches.
📝 Abstract
Agentic AI shifts the investor's role from analytical execution to oversight. We present an agentic strategic asset allocation pipeline in which approximately 50 specialized agents produce capital market assumptions, construct portfolios using over 20 competing methods, and critique and vote on each other's output. A researcher agent proposes new portfolio construction methods not yet represented, and a meta-agent compares past forecasts against realized returns and rewrites agent code and prompts to improve future performance. The entire pipeline is governed by the Investment Policy Statement--the same document that guides human portfolio managers can now constrain and direct autonomous agents.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Agentic AI
Strategic Asset Allocation
Autonomous Portfolio Management
Meta-agent Optimization
Investment Policy Statement
🔎 Similar Papers
No similar papers found.
