This course introduces machine learning methods for econometric analysis and empirical economic research. Students develop R programming skills while mastering statistical learning algorithms in economic contexts.
The curriculum covers supervised learning techniques including predictive regression, k-nearest neighbor algorithms, penalized regression (ridge and lasso), and tree-based methods. Students learn data preprocessing, resampling techniques, cross-validation, and model evaluation. Unsupervised learning topics include k-means clustering and latent Dirichlet allocation (LDA).
The course emphasizes active learning through problem sets and laboratory sessions. Students develop expertise through computational implementations and guided coding sessions rather than traditional lectures.
The course concludes with machine learning applications in causal inference, providing methodological foundations for rigorous empirical economic research.
Prerequisites: basic statistics and elementary R programming.
