Efficient Adaptive Experimental Design for Average Treatment Effect Estimation
read the original abstract
We study how to efficiently estimate average treatment effects (ATEs) using adaptive experiments. In adaptive experiments, experimenters sequentially assign treatments to experimental units while updating treatment assignment probabilities based on past data. We start by defining the efficient treatment-assignment probability, which minimizes the semiparametric efficiency bound for ATE estimation. Our proposed experimental design estimates and uses the efficient treatment-assignment probability to assign treatments. At the end of the proposed design, the experimenter estimates the ATE using a newly proposed Adaptive Augmented Inverse Probability Weighting (A2IPW) estimator. We show that the asymptotic variance of the A2IPW estimator using data from the proposed design achieves the minimized semiparametric efficiency bound. We also analyze the estimator's finite-sample properties and develop nonparametric and nonasymptotic confidence intervals that are valid at any round of the proposed design. These anytime valid confidence intervals allow us to conduct rate-optimal sequential hypothesis testing, allowing for early stopping and reducing necessary sample size.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
Semiparametric Efficiency in Sequential Experiments: Characterization and Design via Average Propensity
Sequential experiments achieve the i.i.d. semiparametric efficiency bound when evaluated at the average propensity induced by the design, and two batched adaptive mechanisms approach this bound with few policy updates.
-
Semiparametric Efficiency in Sequential Experiments: Characterization and Design via Average Propensity
Sequential experiments achieve i.i.d.-level semiparametric efficiency via an induced average propensity score, attained by batched designs using influence-function regression adjustment or adaptive covariate balancing.
-
Active Statistical Inference
Active inference adapts label collection via ML uncertainty to deliver valid statistical inference with substantially fewer samples than standard non-adaptive methods across any data distribution.
-
Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
Transformers trained to imitate Bayesian posterior Neyman allocations achieve smoothness-adaptive ATE estimation via mixture-of-experts in-context learning.
-
Design Stability in Adaptive Experiments: Implications for Treatment Effect Estimation
Derives CLTs and asymptotic variances for IPW and AIPW estimators of ATE under a new design stability condition in adaptive randomization.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.