Taught by me, Kirill Borusyak, at UC Berkeley's Department of Agricultural and Resource Economics, Fall 2023
Please email me at k.borusyak@berkeley.edu if you notice typos, mistakes, or have other suggestions for how to improve the course.
Course outline:
A. Introduction: regression and causality
-
A1. Key facts about regression
-
A2. Potential outcomes and randomized control trials
B. Selection on observables
-
B1. Regression adjustment
-
B2. Matching and propensity score methods
-
B3. Doubly-robust methods
C. Panel data methods
-
C1. Linear panel data methods recap
-
C2. Canonical difference-in-differences (DiD) and event studies
-
C3. DiD with staggered adoption
-
C4. Synthetic control methods and factor models
D. Instrumental variables (IVs)
-
D1. IV idea and mechanics. Weak instruments
-
D2. IV with heterogeneous effects
-
D3. Shift-share IV and formula instruments
-
D4. Examiner designs (“judge IVs”). (+A bit on control functions)
E. Regression discontinuity (RD) designs
-
E1. Sharp RD designs
-
E2. RD extensions: fuzzy RD, spatial RD, RD extrapolation, and more
F. Miscellaneous topics: Models with multiplicative effects and Poisson regression; More on statistical inference
--
Materials are distributed under the Creative Commons Attribution 4.0 license.