Chapter 11: Clarifying derivation of propensity score weighting estimator

matheusfacure / python-causality-handbook

Causal Inference for the Brave and True. A light-hearted yet rigorous approach to learning about impact estimation and causality.

https://matheusfacure.github.io/python-causality-handbook/landing-page.html

MIT License

2.65k stars 463 forks source link

Chapter 11: Clarifying derivation of propensity score weighting estimator #298

Closed loudly-soft closed 1 year ago

loudly-soft commented 1 year ago

In the section "Propensity Weighting" of Chapter 11, I wonder if the second equation:

E[Y * (T-p(x)) / (p(x)(1-p(x))) | X]

could be made more clear in how it relates to the first equation by explicitly stating that:

For T=0, the second equation is equivalent to:

E[-Y / (1-p(x))] * (1-p(x))

For T=1, the second equation is equivalent to:

E[Y / p(x)] * p(x)

The relation to the first equation becomes more clear.