New paper outlines identification and testability of counterfactual maps from covariate heterogeneity
An arXiv paper argues counterfactuals are generally not identified from data alone, since different causal models can yield identical observed distributions but different counterfactuals, and proposes conditions for identification and testability of counterfactual maps via covariate heterogeneity.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-18 05:00 · arXiv stat.ML
Symmetries of Counterfactuals: Identification and Testability of Counterfactual Maps from Covariate Heterogeneity