Important remark: This course is thought simultaneously to Master students. It assumes only basic knowledge of econometrics, and is most appropriate for students who want to get a basic understanding of (mostly) micro-econometric methods used in empirical research.

Does AI reduce job opportunities for young people? Does colonization still explain income differences across countries today? Is there labor market discrimination based on ethnic origin or gender?

Beyond being studied in this course, these diverse questions share a common thread: they require data to separate correlation from causality. In this course, you will study modern econometric techniques to analyze data, evaluate policies, and identify causal effects. You will learn tools such as instrumental variables, panel data techniques, experiments and quasi-experiments.

Given the growing importance of machine learning and AI in redefining how we handle large-scale data, the course also provides an introduction to predictive machine learning techniques as well as time-series forecasting.

Econometrics is best learned by doing. These methods will be illustrated using Stata (and Python for machine learning). Finally, you will bring these skills together in a project where you apply these methods to answer your own research question using a real-world dataset.

The course is based on “Introduction to econometrics” (James Stock, Mark Watson). The introduction to machine learning is based on “Data Analysis for Business, Economics, and Policy” (Gabor Békés, Gébor Kézdi). It covers the following topics:

• Revision of linear regression
• Instrumental variables
• Panel data methods
• Experiments and quasi-experiments (diff-in-diff, regression discontinuity design)
• Introduction to Machine Learning (LASSO, regression trees, random forests)
• Introduction to time-series and forecasting

.