Published in
2 min readMar 26, 2016
Thanks for reaching out, at2014 and others! Following up here with some resources we’ve found useful:
General:
- Mostly Harmless Econometrics by Angrist & Pischke — a must-read practioner’s guide to causal inference.
Controlled Regression (including with Fixed Effects):
- Data Analysis Using Regression and Multilevel/Hierarchical Models by Gelman — see especially Chapters 9 and 10.
Regression Discontinuity Design:
- “Regression Discontinuity Designs: A Guide to Practice” by Imbens and Lemieux in the Journal of Econometrics
- “Identification and Estimation of Treatment Effects with a Regression Discontinuity Design” by Hahn, Todd, and van der Klaauw in Econometrica
Difference-in-Differences
- “Semiparametric Difference-in-Difference Estimators” by Abadie in Review of Economic Studies
Instrumental Variables:
- “Identification of Causal Effects Using Instrumental Variables” by Angrist, Imbens, and Rubin in the Journal of the American Statistical Association
- “Instrumental Variables and the Search for Identification: From Supply and Demand to Natural Experiments” by Angrist and Krueger in the Journal of Economic Perspectives
Extensions:
- “Lectures on Machine Learning” by Athey and Imbens in 2015 NBER Lecture Series — discusses the nascent intersection of machine learning methods with causal inference
- “Something to Talk About: Social Spillovers in Movie Consumption” by Gilchrist and Sands forthcoming in Journal of Political Economy — a paper we wrote in grad school using IV (interacted with a little machine learning) to understand peer effects in consumption
We also put sample R code for these causal methods, along with basic simulation results, in a Jupyter notebook on GitHub. There are a number of helpful R libraries for causal inference, including AER and rdd, that we use there and are worth checking out.
Happy reading!