Method
Causal inference (general)
Effect estimation from observational data: potential outcomes, ATE, quasi-experiments.
12 cases from 10 companies apply Causal inference (general). 1 was checked in an online A/B test; 8 report results in production.
Typical tasks: Measure impact without an A/B test, Decide who gets an offer, Optimize decisions, Balance relevance and diversity and 5 more. Companies: Lyft, Netflix, Databricks, T-Bank, Sber, Khan Academy and 4 more. Often combined with: Bootstrap, Backdoor adjustment, Double machine learning, A/B test, Placebo test and 20 more.
