Advanced Causal Inference with Data

Catalog description

This course is an exploration of the problems that arise in making inferences about causal effects from non-experimental data and some potential solutions to those problems. Topics covered include: regression approaches to causal inference; paradoxes that arise when interpreting non-experimental data; the "potential outcomes" approach to causal inference (often called Rubin's Causal Model); univariate and multivariate matching methods; multivariate distance measures; propensity score estimation methods; using the propensity score to match, subclassify and weight; Instrumental Variables applications of Rubin's Causal Model; Directed Acyclic Graphs.