Lectures
Introduction (slides)
- Topics: statistical learning; parametric vs. nonparametric models; model complexity; training and test errors; bias-variance tradeoff; adaptive statistical models; causal inference; the do-operator; treatment effect; randomized experiment; observational studies; self-selection; causal diagrams; the back-door criterion; quasi-experimental design; regression; matching; instrumental variables; fixed effects; difference-in-differences; regression discontinuity; structural estimation; counterfactual simulation
- Notes and resources: link
codes:R
Part I
Foundations of Statistical Learning (slides)
- Topics: learning theory; VC analysis; approximation-generalization tradeoff; bias-variance tradeoff; information theory; KL divergence; cross entropy; maximum likelihood; decision theory; bayes classifier; regression function; discriminative vs. generative model; scientific model
- Notes and resources: link
Regression (slides)
- Topics: linear regression; OLS estimator; maximum likelihood estimator; Frisch-Waugh-Lovell theorem; gradient descent; hypothesis testing; robust standard error; bootstrap; regression diagnostics; polynomial regression; piecewise constant model; regression splines; generalized additive models
- Notes and resources: link
codes:R
Significance Testing, P-Hacking, and Publication Bias (slides)
- Topics: p-values; multiple testing; Bonferroni correction; publication bias; p-hacking; data-snooping
- Notes and resources: link
Classification and Discrete Choice Models (slides)
- Topics: binary and multiclass classification; generalized linear models; logistic regression; similarity-based methods; K-nearest neighbors (KNN); ROC curve; discrete choice models; random utility maximization (RUM); probit; conditional logit; independence of irrelevant alternatives (IIA)
- Notes and resources: link
codes:R
Model Selection and Regularization (slides)