D0133 - MACHINE LEARNING FOR ASSET MANAGEMENT

Learning outcomes

This course presents a modern view of asset management in which machine learning reshapes the foundations of portfolio construction and financial decision-making. Starting from the limitations of classical approaches based on static assumptions, linear dependencies, and unstable parameter estimation, the course introduces data-driven methods that adapt to evolving market regimes and complex cross-asset interactions. Students will study the full asset management process—from information extraction and signal aggregation to dynamic asset allocation and risk control—through the lens of machine learning. Emphasis is placed on robustness, stability, and economic interpretability rather than purely predictive accuracy. The course integrates recent advances from academic research and industry practice, showing how machine learning enables flexible allocation frameworks, improved diversification and more realistic handling of uncertainty. By the end of the course, students will understand how modern ML-based asset management departs from traditional portfolio theory and opens new directions for quantitative investment design

Course contents

This course presents a modern view of asset management in which machine learning reshapes the foundations of portfolio construction and financial decision-making. Starting from the limitations of classical approaches based on static assumptions, linear dependencies, and unstable parameter estimation, the course introduces data-driven methods that adapt to evolving market regimes and complex cross-asset interactions. Students will study the full asset management process—from information extraction and signal aggregation to dynamic asset allocation and risk control—through the lens of machine learning. Emphasis is placed on robustness, stability, and economic interpretability rather than purely predictive accuracy. The course integrates recent advances from academic research and industry practice, showing how machine learning enables flexible allocation frameworks, improved diversification and more realistic handling of uncertainty. By the end of the course, students will understand how modern ML-based asset management departs from traditional portfolio theory and opens new directions for quantitative investment design.

Readings/Bibliography

• López de Prado, M. (2018). Advances in Financial Machine Learning. Wiley

• López de Prado, M. (2020). Machine Learning for Asset Managers. Cambridge University Press

• López de Prado, M. (2023). Causal Factor Investing. Cambridge University Press (Elements in Quantitative Finance).

• Guillaume Coqueret & Tony Guida (2023). Machine Learning for Factor Investing: Python Version (CRC Press)

• Jansen, S. (2026). Machine Learning for Trading (3rd ed.).

• Hamlet Jesse Medina Ruiz & Ernest P. Chan (2025). Generative AI for Trading and Asset Management (Wiley)

Office hours

See the website of Maurizio Morini