Proposes nonparametric tests for treatment effect heterogeneity that avoid sample splitting, incorporate structured assumptions, and target policy-relevant alternatives comparing personalized vs. covariate-ignoring rules.
Let t+ α and t− α be chosen as the (1 − α) quantile of supf ∈F G+(f) and the α quantile of inf f ∈F G−(f) respectively
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Nonparametric tests of treatment effect homogeneity for policy-makers
Proposes nonparametric tests for treatment effect heterogeneity that avoid sample splitting, incorporate structured assumptions, and target policy-relevant alternatives comparing personalized vs. covariate-ignoring rules.