Loss-weighted targeting in TMLE introduces more systematic bias than clever-covariate-scaled targeting under positivity stress, while a proposed Lepski-type adaptive truncation with brake improves stability over fixed rules like c/(sqrt(n) log n) with c=5 or 6.
arXiv preprint arXiv:2405.07186 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
verdicts
UNVERDICTED 3roles
method 1polarities
use method 1representative citing papers
The paper formalizes identification strategies for potential outcome means and average treatment effects when merging experimental studies with external data sources.
The adaptive influence-based borrowing framework selects subsets of external controls by influence scores and chooses the subset minimizing MSE of the treatment effect estimator.
citing papers explorer
-
Investigating Targeting Strategies and Truncation in TMLE for the Average Treatment Effect under Practical Positivity Violations
Loss-weighted targeting in TMLE introduces more systematic bias than clever-covariate-scaled targeting under positivity stress, while a proposed Lepski-type adaptive truncation with brake improves stability over fixed rules like c/(sqrt(n) log n) with c=5 or 6.
-
Identification strategies for combining an experimental study with external data
The paper formalizes identification strategies for potential outcome means and average treatment effects when merging experimental studies with external data sources.
-
Adaptive Influence-Based Borrowing Framework for Improving Treatment Effect Estimation in RCTs Using External Controls
The adaptive influence-based borrowing framework selects subsets of external controls by influence scores and chooses the subset minimizing MSE of the treatment effect estimator.