Part III · Advanced Causal Machine Learning
Module 09. Instrumental Variables & the LATE
Use external variation to beat unobserved confounding, and be precise about whom the estimate describes.
Education economics: distance to college and the exclusion restriction about 6 hours3 lessons
Module progress0/3 lessons · 0%
Lessons
9.1
Instruments, the Wald Estimator and Compliers
- · State the three IV conditions: relevance, independence and the exclusion restriction
- · Compute the Wald estimator as reduced form divided by first stage
- · Classify units as compliers, always-takers, never-takers and defiers
120 min 5 Colab cellsVideo coming soon
9.2
Two-Stage Least Squares and Weak Instruments
- · Estimate 2SLS by hand and with linearmodels, and see why manual second stages get the standard errors wrong
- · Estimate the return to schooling with college proximity as an instrument (Card, 1995)
- · Read first-stage diagnostics and apply the F > 10 rule of thumb (and its modern caveats)
120 min 5 Colab cellsVideo coming soon
9.3
The Exclusion Restriction, Endogeneity Tests and Over-Identification
- · Simulate violations of the exclusion restriction and see how far they move the estimate
- · Run and interpret the Wu–Hausman test of endogeneity
- · Use two instruments and interpret the Sargan over-identification test, including its limits
120 min 5 Colab cellsVideo coming soon