How deeply does Causal Inference for Data Science cover doubly robust estimation?

Explains AIPW as the primary doubly robust method, combining T-learner and IPW rather than surveying every DR technique.

Doubly robust estimation

As explained in Causal Inference for Data Science

A doubly robust estimator is a causal parameter estimator that combines two estimators and can remain consistent when at least one underlying model is correctly specified.

Doubly robust estimation improves causal-effect estimates by combining two estimators. If either underlying estimator is unbiased, the combined estimator remains unbiased, providing a safety net when one model is misspecified. It is intended for situations where accuracy and precision matter, such as healthcare or finance. section 8.3 section 8.0

Combining outcome and treatment models

Suppose T is a binary treatment, Y is an outcome, and C contains the confounders. A T-learner estimates expected outcomes under treatment and control, while inverse probability weighting (IPW) estimates propensity scores: the probability of treatment given C. Augmented inverse probability weighting (AIPW) combines these two approaches into a doubly robust estimator of the average treatment effect. The setup assumes no unobserved confounders. section 8.3 section 8.3

Table 8.4 Available data with propensity scores calculated

Y	T	C ₁	…	C_p
y ₁	t ¹	c ¹ ₁	…	c ¹ _p
…	…	…	…	…
y _n	t ⁿ	c ⁿ ₁	…	c ⁿ _p

Inspect the available outcome, treatment, confounder, and propensity-score data used to construct the estimator. section 8.3

How Causal Inference for Data Science covers it

Cover of Causal Inference for Data Science

liveBook manning.com

Read Causal Inference for Data Science on liveBook Get the book at manning.com

Related concepts

DoubleMLimplemented by · DoubleML is an open-source software library for estimating causal effects with double machine learning and related methods in R and Python.
augmented inverse probability weightingincludes · Augmented inverse probability weighting is a doubly robust causal estimation method that combines an outcome model with inverse-probability treatment weighting.

This explainer was generated by Manning's knowledge graph from the text of Causal Inference for Data Science. The section links point to the parts of the book it draws on.