DML confidence-interval coverage is highly sensitive to the nuisance learner, with rates from 0% to 100% across simulations, and bootstrap intervals do not consistently fix under-coverage.
Bootstrap vs Asymptotic Variance Estimation When Using Propensity Score Weighting with Continuous and Binary Outcomes
1 Pith paper cite this work, alongside 43 external citations. Polarity classification is still indexing.
1
Pith paper citing it
43
external citations · OpenAlex
fields
stat.ML 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Analytical and Bootstrap Confidence Intervals of Double Machine Learning: Simulation studies and an application to rural-urban difference in obesity prevalence
DML confidence-interval coverage is highly sensitive to the nuisance learner, with rates from 0% to 100% across simulations, and bootstrap intervals do not consistently fix under-coverage.