The following is tentative schedule for the semester that is
very much in flux. For each week, I offer readings that
I believe will help you understand the material we will cover on a
deeper level. You are strongly recommended to read these ahead of our
class meeting.
Module 1:
Introduction and Potential Outcomes (Sept 1-8)
Topics
- Potential outcomes
- Causal estimands
- Introduction to causal graphs
Reading
- Imbens & Rubin, Ch. 1
- Angrist & Pischke, Chapter 1.
- Holland, P. W. 1986.
“Statistics and Causal
Inference”. Journal of the American Statistical Association, Vol.
81, No. 396: 945-960.
Module 2:
Randomization Inference (Sept 13-15)
Topics
- Randomized experiments
- Fisher’s approach to inference, permutation tests
- Sharp null, randomization distribution
Reading
- Imbens and Rubin, Chapter 5. (Skim Chapter 4 for some
definitions.)
- Rosenbaum, Paul R. 2002. Observational Studies.
Springer-Verlag. 2nd edition. Chapter 2.
Module
3: Inference for the Average Treatment Effect (Sept 20-22)
Topics
- Neyman’s approach to inference for the ATE
- Finite-sample vs superpopulation inference
- Stratified and matched-pair randomized trials
Reading
- Imbens & Rubin, Chapters 6, 9 (Skip 9.6–9.7), and 10 (Skip
10.6–10.7)
- Angrist and Pischke: Chapter 2.
Module
4: Linear Regression and Randomized Experiments (Sept 27-29)
Topics
- Simple linear regression in experiments
- Cluster-randomized trials
- Covariate adjustment in experiments with regression
Reading
Module 5: Observational
Studies (Oct 4-13)
Topics
- Selection on observables,
- DAGs and the back-door criterion
- Regression for observational data
- Sensitivity analysis and partial identification
Reading
- Angrist & Pischke, Chapter 3.
- Imbens & Rubin, Chapters 21 and 22.
- Morgan & Winship, Chapter 4.
Module 6:
Instrumental Variables (Oct 18-20)
Topics
- Noncompliance in randomized experiments
- Two-stage least squares
- Instrumental variables in observational studies
Reading
- Imbens & Rubin, Chapters 23 and 24
- Angrist & Pischke, Chapter 4
Module 7:
Matching and Weighting Estimators (Oct 25-27)
Topics
- Propensity scores, matching and weighting estimators
- Optimal matching methods
- Calibration methods
Reading
- Imbens & Rubin, Chapters 13, 15, and 18.
- Stuart. 2010. . Statistical Science. Vol. 25, No. 1: 1–21
Module 8:
Regression Discontinuity Designs (Nov 1-3)
Topics
- Sharp RD designs, identification
- Estimation and bandwidth selection
- Fuzzy RD designs and diagnostics
Reading
Module
9: Panel Data, Fixed Effects, and Differences-in-differences (Nov
8-15)
Topics
- Fixed effects and first differences
- Difference-in-differences
- Synthetic control methods
Reading
Topics
- Causal heterogeneity and effect modification
- Causal mediation and direct effects
Reading
Module 11: Final topics
(Nov 30-Dec 1)
Topics
- Time-varying treatments and marginal structural models
- Exploring treatment effect heterogeneity
Reading