Causal Identification from Counterfactual Data: Completeness and Bounding Results

Researchers developed the CTFIDU+ algorithm for causal identification using counterfactual data, establishing theoretical limits for exact causal inference in non-parametric settings. The work extends previous completeness results by incorporating Layer 3 counterfactual distributions that can be experimentally obtained, and provides novel bounds for non-identifiable quantities.

• →The CTFIDU+ algorithm enables identification of counterfactual queries from arbitrary Layer 3 distributions with proven completeness.
• →Research establishes the fundamental theoretical limit for exact causal inference in non-parametric settings using physically realizable distributions.
• →Novel analytic bounds are derived for critical counterfactuals that cannot be identified, with counterfactual data helping tighten these bounds.
• →Work builds on recent advances in counterfactual realizability that allow experimental estimation of certain Layer 3 distributions.
• →Results extend beyond traditional observational and interventional data to include previously inaccessible counterfactual distributions.