Econ 224 - Statistical Learning and Causal Inference for Economics

Econ 224 is a new course that will be offered for the first time in Fall 2018. The only pre-requisite is Econ 103, Introductory Statistics for Economists. Econ 224 will mainly be based on two books:

The first part of the course will cover prediction and statistical learning, including linear regression, ridge regression, LASSO, logistic regression, regression trees, and random forests. The second part of the course will cover causal inference: potential outcomes, randomized controlled trials, regression under unconfoundedness, isntrumental variables, regression discontinuity, and differences-in-differences. The final part of the course will examine how economists are increasingly combining tools from statistical learning and causal inference to tackle exciting policy questions.

Since Econ 104 is not a pre-requisite, there will unavoidably be some overlap between 224 and 104. But even when they treat related material, the emphasis will be very different. In particular, Econ 224 will be a very applied class, and will make heavy use of R. Grades will mainly be based on applied data analysis problem sets and projects rather than exams. By the end of the course, you’ll be able to applied fairly sophisticated machine learning and causal inference tools on important problems.

Watch this space for further updates, including a syllabus, in the coming months.