Causal Inference & Machine Learning Course

This is a book that contains the labs and reports I elaborated for the Causal Inference & Machine Learning course I took at Universidad del Pacifico.

Note

The labs were a collaborative effort involving Valerie Dube, Erzo Garay, Juan Guerrero, and myself. We developed three versions of the labs, implemented in Python, R, and Julia.

The course, taught by Professor Alexander Quispe Rojas, bridges the gap between causal inference in economic analysis and machine learning methods. It covers classical econometric structural models and their modern artificial intelligence counterparts, known as Directed Acyclic Graphs (DAGs). Throughout the course, we explored various machine learning and AI tools (lasso, random forest, and deep neural networks) for inferring causal parameters and quantifying uncertainty.

Clic here for the complete syllabus.

Labs

• 1. Intro to Partialling-out and Lasso Cross-validation
• 2. Multicollinearity, Precision Adjustment, and DAGs for Good and Bad controls
• 3. Neyman Orthogonality, Orthogonal Learning, and Double Lasso
• 4. Bootstraping and Causal Forest Analysis
• 5. Double Machine Learning

Reports

• Report 1
• Report 2
• Report 3
• Report 4
• Report 5