Course · Advanced - Machine Learning for Market Research

Learning outcomes

• Design predictive features from market and fundamental data
• Avoid temporal information leakage when validating models on financial series
• Train and compare ensemble models for classification and regression tasks
• Translate a model's predictions into portfolio construction rules and risk analysis

The closing module of the curriculum, aimed at anyone who already has a solid grasp of basic quantitative pricing and wants to rigorously bring machine learning into their analysis.

Who it's for

Anyone with prior experience in Python and statistics who wants to apply machine learning models to financial data without falling into the most common methodological mistakes in this field.

What the module covers

• Feature engineering from prices, fundamentals and macro data
• Temporal validation schemes that prevent information leaking from the past
• Ensemble models (random forests, boosting) applied to financial series
• From prediction to portfolio: construction rules and risk attribution

Format

Recorded sessions with complete practical cases, from raw data to the evaluation of a strategy based on the trained model.