Two debiased estimators, one based on linear programming solutions and one on entropic smoothing, provide asymptotic confidence intervals for covariate-dependent partial identification bounds and support policy learning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Partial identification via conditional linear programs: estimation and policy learning
Two debiased estimators, one based on linear programming solutions and one on entropic smoothing, provide asymptotic confidence intervals for covariate-dependent partial identification bounds and support policy learning.