Machine Learning for Cross-Sectional Multi-Factor Strategies

Summary

This announcement outlines a two-day VeighNa training course on machine-learning cross-sectional multi-factor strategies, also described as alpha strategies. It presents their use in equity selection, index enhancement, absolute-return portfolios, and leveraged long-short applications involving futures and other assets. The curriculum covers factor data, predictive targets, model training and validation, signal evaluation, portfolio construction, and backtesting details. Named examples include LightGBM, Lasso, XGBoost, and LSTM.

The course description emphasizes the full research workflow: preparing and cleaning features, assessing predictive signals and feature importance, tuning models, and translating cross-sectional forecasts into portfolios. It also mentions evaluating factors with Alphalens and exploring intraday alpha features. The document is a course promotion, not a strategy paper: it presents no independent performance evidence, and it does not provide the course materials or enough methodological detail to reproduce a trading system. It cautions that the topic is complex and is not recommended for beginners.

Key ideas

• Cross-sectional multi-factor strategies use features to compare assets and generate relative alpha signals.
• The course applies the approach to equities and leveraged long-short portfolios across several asset types.
• Its curriculum spans factor data preparation, model training, validation, interpretability, and portfolio construction.
• Model examples include LightGBM, Lasso, XGBoost, and LSTM.
• The announcement provides a syllabus but no strategy results or reproducible implementation details.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.