Causal Inference with Spatial Treatments
Two ideas on how to do causal inference when treatment is a point in space
On the use and misuse of Fisher Randomization Tests
Randomly assigning treatment or treatment dates do not show observational estimators are robust (except in special cases)
Multiplying matrices; order maters
A quick example of why you should care about matrix dimensions
A note on did2s syntax
I get a lot of questions about `first_stage` vs. `second_stage` in `did2s`. This note hopefully will help clarify it
Factor Models and Synthetic Control
Understanding factor models is important for understanding (advancements in) the synthetic control method.
Difference-in-Differences: Unit vs. Group FEs
In difference-in-differences settings, should you allow for unit-level fixed effects or just treatment-cohort (group) fixed effects? Jeff Wooldridge's paper shows when you have a balanced paper, imputation estimators produce the same, but group fixed effects can be *way* faster.
Non-traditional Diff-in-Diff
Notes on how I think about non-traditional difference-in-differences settings. I first describe the intuition behind the basic two-stage diff-in-diff method and then use that insight to describe how to approach non-traditional settings (e.g. continuous treatment, treatment turning on and off, and multiple doses during period)
All about influence functions
Introductory notes with detailed derivations of influence functions. I took these notes while trying to learn the material myself.