Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy
From MaRDI portal
Publication:282469
DOI10.1214/15-AOS1384zbMATH Open1338.62089arXiv1603.07573WikidataQ91130559 ScholiaQ91130559MaRDI QIDQ282469FDOQ282469
Authors: Alexander R. Luedtke, Mark J. Van der Laan 
Publication date: 12 May 2016
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: We consider challenges that arise in the estimation of the mean outcome under an optimal individualized treatment strategy defined as the treatment rule that maximizes the population mean outcome, where the candidate treatment rules are restricted to depend on baseline covariates. We prove a necessary and sufficient condition for the pathwise differentiability of the optimal value, a key condition needed to develop a regular and asymptotically linear (RAL) estimator of the optimal value. The stated condition is slightly more general than the previous condition implied in the literature. We then describe an approach to obtain root- rate confidence intervals for the optimal value even when the parameter is not pathwise differentiable. We provide conditions under which our estimator is RAL and asymptotically efficient when the mean outcome is pathwise differentiable. We also outline an extension of our approach to a multiple time point problem. All of our results are supported by simulations.
Full work available at URL: https://arxiv.org/abs/1603.07573
Recommendations
- Targeted sequential design for targeted learning inference of the optimal treatment rule and its mean reward
- Statistical learning of origin-specific statically optimal individualized treatment rules
- Resampling‐based confidence intervals for model‐free robust inference on optimal treatment regimes
- Improved doubly robust estimation in learning optimal individualized treatment rules
- Quantile-optimal treatment regimes
semiparametric modelonline estimationefficient estimatornon-regular inferenceoptimal treatmentoptimal valuepathwise differentiability
Cites Work
- Weak convergence and empirical processes. With applications to statistics
- Measure Theory and Probability Theory
- Title not available (Why is that?)
- Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy
- Estimating individualized treatment rules using outcome weighted learning
- Performance guarantees for individualized treatment rules
- A robust method for estimating optimal treatment regimes
- Optimal Structural Nested Models for Optimal Sequential Decisions
- Adaptive confidence intervals for the test error in classification
- Optimal aggregation of classifiers in statistical learning.
- Nonexistence of informative unbiased estimators in singular problems
- Impossibility results for nondifferentiable functionals
- Sparse online learning via truncated gradient
- Statistical methods for dynamic treatment regimes. Reinforcement learning, causal inference, and personalized medicine
- Dynamic treatment regimes: technical challenges and applications
- Discussion of ``Dynamic treatment regimes: technical challenges and applications
- Comment on ``Dynamic treatment regimes: technical challenges and applications
- Rejoinder of ``Dynamic treatment regimes: technical challenges and applications
- On central limit theorems for martingale triangular arrays
- Title not available (Why is that?)
- Fast learning rates for plug-in classifiers
Cited In (49)
- Optimal Individualized Decision Rules Using Instrumental Variable Methods
- Targeted sequential design for targeted learning inference of the optimal treatment rule and its mean reward
- A multiagent reinforcement learning framework for off-policy evaluation in two-sided markets
- Rejoinder: The future of outcome-wide studies
- Transformation-Invariant Learning of Optimal Individualized Decision Rules with Time-to-Event Outcomes
- High-dimensional \(A\)-learning for optimal dynamic treatment regimes
- Off-Policy Confidence Interval Estimation with Confounded Markov Decision Process
- Statistically Efficient Advantage Learning for Offline Reinforcement Learning in Infinite Horizons
- Performance guarantees for policy learning
- Asymptotic Properties of Stationary Solutions of Coupled Nonconvex Nonsmooth Empirical Risk Minimization
- Relative contrast estimation and inference for treatment recommendation
- Statistical Inference for High-Dimensional Models via Recursive Online-Score Estimation
- Parametric-rate inference for one-sided differentiable parameters
- On testing conditional qualitative treatment effects
- Learning optimal personalized treatment rules in consideration of benefit and risk: with an application to treating type 2 diabetes patients with insulin therapies
- Concordance and value information criteria for optimal treatment decision
- Transfer learning for contextual multi-armed bandits
- A non-parametric Bayesian approach for adjusting partial compliance in sequential decision making
- Estimation and evaluation of individualized treatment rules following multiple imputation
- Robust method for optimal treatment decision making based on survival data
- Doubly robust treatment effect estimation with missing attributes
- Inferences on the difference between future observations for comparing two treatments
- Sensitivity Analysis via the Proportion of Unmeasured Confounding
- A General Framework for Subgroup Detection via One-Step Value Difference Estimation
- Value Enhancement of Reinforcement Learning via Efficient and Robust Trust Region Optimization
- Subgroup identification and variable selection for treatment decision making
- Estimators for the value of the optimal dynamic treatment rule with application to criminal justice interventions
- The optimal dynamic treatment rule superlearner: considerations, performance, and application to criminal justice interventions
- Correction to: ``Statistical inference for the mean outcome under a possibly nonunique optimal treatment rule
- Doubly Robust Interval Estimation for Optimal Policy Evaluation in Online Learning
- A Sparse Random Projection-Based Test for Overall Qualitative Treatment Effects
- Title not available (Why is that?)
- Modern approaches for evaluating treatment effect heterogeneity from clinical trials and observational data
- Statistical inference for online decision making via stochastic gradient descent
- Estimation of Conditional Average Treatment Effects With High-Dimensional Data
- Finite-sample optimal estimation and inference on average treatment effects under unconfoundedness
- A Generalized Estimator of the Attributable Benefit of an Optimal Treatment Regime
- Experimental Evaluation of Individualized Treatment Rules
- Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy
- Model-Assisted Uniformly Honest Inference for Optimal Treatment Regimes in High Dimension
- Resampling‐based confidence intervals for model‐free robust inference on optimal treatment regimes
- CSTE curve for selection of the optimal treatment when outcome is binary
- Inference for non-regular parameters in optimal dynamic treatment regimes
- Sharp instruments for classifying compliers and generalizing causal effects
- Title not available (Why is that?)
- Proper inference for value function in high-dimensional Q-learning for dynamic treatment regimes
- Deep spectral Q-learning with application to mobile health
- Rejoinder: Optimal Individualized Decision Rules Using Instrumental Variable Methods
- Optimal Treatment Regimes: A Review and Empirical Comparison
Uses Software
This page was built for publication: Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q282469)
