A single scalar rank-stickiness parameter nonparametrically point-identifies the entire treatment-effect distribution by selecting the Bregman-Sinkhorn copula that maximizes average rank correlation subject to a relative-entropy constraint.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Develops novel one-step and TMLE estimators for ATE and ATT under front-door assumptions with ML nuisance estimation, root-n consistency proofs, and doubly robust tests for identification assumptions.
citing papers explorer
-
Nonparametric Point Identification of Treatment Effect Distributions via Rank Stickiness
A single scalar rank-stickiness parameter nonparametrically point-identifies the entire treatment-effect distribution by selecting the Bregman-Sinkhorn copula that maximizes average rank correlation subject to a relative-entropy constraint.
-
Flexible Nonparametric Inference for Causal Effects under the Front-Door Model
Develops novel one-step and TMLE estimators for ATE and ATT under front-door assumptions with ML nuisance estimation, root-n consistency proofs, and doubly robust tests for identification assumptions.