Causal Inference

Undergraduate / Graduate

This course reviews modern causal inference techniques and their applications in business and economics. The course covers the treatment-control comparison estimator, regression adjustment, matching (on covariates and propensity score), difference in differences (canonical and with staggered treatment), panel data methods, regression discontinuity design (sharp and fuzzy), instrumental variables and local average treatment effect (LATE) estimator. At different points during the course, we mention how machine learning (ML) techniques have recently been used to enrich the classical methods. The course involves programming (R language) and working with data. Students are expected to have a solid background in statistics (working knowledge of R and familiarity with RStudio) and econometrics.