Causal Inference 101

4 min read

Doubly robust estimation

Why combining propensity and outcome models can protect you when one model is slightly wrong.

Doubly robust (DR) estimators use both a propensity model and an outcome model. If either is correctly specified (not necessarily both), the treatment effect estimate can be consistent. That is valuable in high-dimensional settings.

DR-learner and related methods in the EconML family implement this logic with machine learning for the nuisance models.

Run this method on your data — no Python

CausalLens runs matching, DiD, causal forests, DoWhy refutation, and more — with balance tables, sensitivity checks, and PDF export.