A general characterization of optimal tie-breaker designs

DOI10.1214/23-AOS2275arXiv2202.12511OpenAlexW4386035723MaRDI QIDQ6136581FDOQ6136581

Authors: Harrison H. Li, Art B. Owen 

Publication date: 31 August 2023

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Tie-breaker designs trade off a statistical design objective with short-term gain from preferentially assigning a binary treatment to those with high values of a running variable $x$. The design objective is any continuous function of the expected information matrix in a two-line regression model, and short-term gain is expressed as the covariance between the running variable and the treatment indicator. We investigate how to specify design functions indicating treatment probabilities as a function of $x$ to optimize these competing objectives, under external constraints on the number of subjects receiving treatment. Our results include sharp existence and uniqueness guarantees, while accommodating the ethically appealing requirement that treatment probabilities are non-decreasing in $x$. Under such a constraint, there always exists an optimal design function that is constant below and above a single discontinuity. When the running variable distribution is not symmetric or the fraction of subjects receiving the treatment is not $1/2$, our optimal designs improve upon a $D$-optimality objective without sacrificing short-term gain, compared to the three level tie-breaker designs of Owen and Varian (2020) that fix treatment probabilities at $0$, $1/2$, and $1$. We illustrate our optimal designs with data from Head Start, an early childhood government intervention program.

Full work available at URL: https://arxiv.org/abs/2202.12511

Recommendations

• Optimal designs for stated choice experiments that incorporate ties
• Optimality of a class of efficiency-balanced designs
• Optimal tie-breaking rules
• A note on optimal bounded designs
• scientific article; zbMATH DE number 63479
• Optimality of Some Class of Incomplete Block Designs

zbMATH Keywords

D-optimalityregression discontinuityequivalence theoremconstrained optimal designcompound designcutoff design

Mathematics Subject Classification ID

Optimal statistical designs (62K05)

Cites Work

• Title not available (Why is that?)
• Testing Statistical Hypotheses
• Locally Optimal Designs for Estimating Parameters
• Selecting a biased-coin design
• Asymptotic properties of covariate-adjusted response-adaptive designs
• Optimum biased coin designs for sequential clinical trials with prognostic factors
• The Theory of Response‐Adaptive Randomization in Clinical Trials
• Forcing a sequential experiment to be balanced
• Handling covariates in the design of clinical trials
• On the Equivalence of Constrained and Compound Optimal Designs
• Selection bias in evaluating treatment effects: some formal illustrations
• Experimental design in a class of models
• Title not available (Why is that?)
• Constrained optimal designs
• Optimal multipurpose designs for regression models
• Constrained optimal designs for regressiom models
• Optimal Experimental Design for Polynomial Regression
• Title not available (Why is that?)
• Adaptive designs for normal responses with prognostic factors
• The Equivalence of Constrained and Weighted Designs in Multiple Objective Design Problems
• On the Fundamental Lemma of Neyman and Pearson
• A new family of covariate-adjusted response adaptive designs and their properties
• Information-regret compromise in covariate-adaptive treatment allocation
• A unified family of covariate-adjusted response-adaptive designs based on efficiency and ethics
• A linear response bandit problem
• The implicit function theorem for maps that are only differentiable: an elementary proof
• Research design meets market design: using centralized assignment for impact evaluation
• Optimizing the tie-breaker regression discontinuity design
• Kernel regression analysis of tie-breaker designs
• Marginal effects of merit aid for low-income students

This page was built for publication: A general characterization of optimal tie-breaker designs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136581)