New worst-case randomization tests yield valid p-values for sharp and bounded null hypotheses on treatment effects under potentially informative sample attrition by incorporating potential missingness indicators and allowing structural assumptions like monotone missingness.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
econ.EM 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Randomization Inference with Sample Attrition
New worst-case randomization tests yield valid p-values for sharp and bounded null hypotheses on treatment effects under potentially informative sample attrition by incorporating potential missingness indicators and allowing structural assumptions like monotone missingness.