Scientific inquiry is dependent on progress made in theory, measurement, observation and causal inference. This class will review the fundamental and contemporary theories of causal inference as found in Neyman, Rubin and Pearl. Such theories are crucial both in scientific inquiry and policy design. These theories define causality in terms of comparing historical events to their counterfactuals. We will review several research designs that can under varied assumptions and with appropriate statistical models estimate these causal effects credibly. The class will have prescribed assignments in R as the secondary objective of the course is to help students gain competency in the implementation of these designs as well as understanding them more generally.
This class will consist of all-day lectures and exercises, from 9am to 5pm Texas time each class day. Class days are Monday, Wednesday, and Friday in week 1, and Monday and Wednesday in week 2.
Read this website as preparation: https://www.tellingstorieswithdata.com/
Textbook: Causal Inference: The Mixtape, written by the professor.
Our favorite causal inference guru Prof. Scott Cunningham is teaching a free class on causal inference for students from low- and middle-income countries. We're now looking for TAs for this class! The work is condensed into a few days in July (see dates), and will involve some live teaching and some grading.
You have to be a PhD student, or hold a PhD, in economics or a related discipline, and have a solid grasp of econometric methods, especially causal inference. Ideally you've previously taught/TA'd for an econometrics class. When you sign up, please describe in detail whether and how you meet these criteria.
Important: In contrast to the rest of this website, you do NOT have to be a citizen/resident of a low- or middle-income country to be a TA for this class.