Course Description
This course explores modern topics in causal inference with a focus on randomized experiments and experimental design. The course develops the potential outcomes framework and builds toward advanced topics including covariate balancing, overlap conditions and positivity violations, causal inference under interference, and policy evaluation. The emphasis is on both theoretical foundations and applied practice. Causal inference from observational data is de-emphasized.
Students will complete presentation and project work including both applied and theoretical replication studies.
Instructors
Instructor: Prof. Johan Ugander (johan.ugander@)
Schedule
Class W 9:25a-11:20a, location TBD
Office Hours Johan: TBD
Important Links
Course structure and evaluation
The class is organized around one meeting per week. Most of these meetings will be scaffolded by a lecture format, but the purpose of the class is to actively debate methodological choices in causal inference, so active participation will be expected. Attendance is mandatory, and please email me in advance of any meetings that you think you may miss.
There will be two smaller problem sets. The focus of the coursework will be on three projects for the course, with student presentations: (i) replications of a “method paper”, (ii) replications of an “applied paper”, and (iii) an open-ended final project. The exact timeline and format of the projects will be determined in the first week of class, pending class size.
During Weeks 1 & 2 (before the “Drop date,” Friday Sept 11), all students will be required to meet in person (1-1) with Prof. Ugander for a 10 minutes as a “check-in”. This is a PhD level course and I’d like to get to know everyone. Projects will also require check-in meetings.
Completing the assignments may require setting up accounts and using APIs that aren’t always free, or where paid versions are far superior to free versions. I constantly keep an eye out for free alternatives, but also want you to train with the best tools of today. Therefore I ask students be prepared to spend approximately $30-60 as part of this class. If this would be a hardship then please reach out to me. There are no required textbooks or other course materials.
Grading and late policies
- 20% Problem Sets (10% each)
- 60% Projects (20% each)
- 20% Attendance, participation and collegiality
On deadlines: your best bet is to hit all the deadlines. Failing that, the following policy is designed to prioritize completing problem sets over not completing them, submitting problem sets over not submitting them.
- One single delay of up to 2 days (48 hours), no questions asked.
- After that: 5 points deducted per day (per 24h, including weekends), up to 6 days.
- After that: Max grade is a 70. Must be submitted by 11:59p on the final day of semester (after that, grade is 0), and no guarantees on grading turn-around time.
Project: No extensions on the projects, as they require student presentations that need to be scheduled carefully. If you need an extension, please contact me to discuss my policy for taking an incomplete for the course.
In the event of a family or medical emergency, I am fundamentally a reasonable person. Please contact me as soon as possible.
Academic integrity
Academic integrity is a core institutional value at Yale. It means, among other things, truth in presentation, diligence and precision in citing works and ideas we have used, and acknowledging our collaborations with others.
Collaboration in this course is strongly encouraged, but all collaboration must be acknowledged. Failure to disclose collaboration (including allowing someone else to represent your work as their own) will be considered a violation of academic integrity and will be reported to Yale College for arbitration.
AI Policy
I strongly encourage you to use and sharpen your use of AI tools as part of this course. For each problem set and for the projects, I will ask you to document how you used AI, what was useful and what wasn’t, so we can all learn together about best practices. Mindless use of AI (i.e., submitting work that is nonsensical or wrong in ways where you can’t explain what you did) will be graded harshly. Undocumented use of AI will be considered a violation of academic integrity.
High level course schedule
| Lecture | Date | Topic | Notes | |
|---|---|---|---|---|
| L01 | 9/2 | Introduction | ||
| L02 | 9/9 | Basic estimators, common pitfalls | ||
| L03 | 9/16 | Covariate balancing, experimental design | ||
| L04 | 9/23 | Positivity, trimming, target populations, partial identification | ||
| *P01 | 9/30 | Presentations I* | ||
| L05 | 10/7 | Interference | Project I Due | |
| L06 | 10/14 | Estimation with structural models | ||
| Fall break | ||||
| L07 | 10/28 | Policy evaluation; panel data | ||
| *P02 | 11/4 | Presentations II* | ||
| L08 | 11/11 | Advanced estimation topics | Project II Due | |
| L09 | 11/18 | AI for causal inference | ||
| Thanksgiving break | ||||
| L10 | 12/2 | Overflow | ||
| *P03 | 12/9 | Presentations III* | ||
| Exam week | (No exam) | Project III Due |
Accommodations
Yale University is committed to providing equal access to its academic programs. Students who may need an academic accommodation based on the impact of a disability should contact Student Accessibility Services (SAS) to initiate the process. SAS will review appropriate documentation, determine eligibility, and issue an official accommodation letter.
If you are approved for accommodations for this course (e.g., extended exam time or other adjustments), please email me as early as possible and include your accommodation letter from SAS so that we can make appropriate arrangements in a timely manner. Early communication is important to ensure that accommodations can be implemented effectively.
For more information, please visit: https://sas.yale.edu
Computing Environment
A Unix-like setup is strongly recommended (e.g., Linux, Mac OS X, or Cygwin). We will use Python 3 (JupyterLab is recommended).