Biometrika
Feb 16, 2026
Randomization-Based Confidence Sets for the Local Average Treatment Effect
Authors
P M Aronow
Haoge Chang
Patrick Lopatto
Research Topics
Causal Inference
Experimental Design
Paper Information
-
Journal:
Biometrika -
DOI:
10.1093/biomet/asag010 -
Published:
February 16, 2026 -
Added to Tracker:
Feb 17, 2026
Abstract
Summary We consider the problem of generating confidence sets in randomized experiments with noncompliance. We show that a refinement of a randomization-based procedure proposed by Imbens & Rosenbaum (2005) has desirable properties. Specifically, we show that using a studentized Anderson–Rubin statistic as a test statistic yields confidence sets that are finite-sample exact under treatment effect homogeneity and remain asymptotically valid for the local average treatment effect when the treatment effects are heterogeneous. We provide a uniform analysis of this procedure and efficient algorithms to construct the confidence sets.
Author Details
P M Aronow
AuthorHaoge Chang
AuthorPatrick Lopatto
AuthorResearch Topics & Keywords
Causal Inference
Research AreaExperimental Design
Research AreaCitation Information
APA Format
P M Aronow
,
Haoge Chang
&
Patrick Lopatto
(2026)
.
Randomization-Based Confidence Sets for the Local Average Treatment Effect.
Biometrika
, 10.1093/biomet/asag010.
BibTeX Format
@article{paper907,
title = { Randomization-Based Confidence Sets for the Local Average Treatment Effect },
author = {
P M Aronow
and Haoge Chang
and Patrick Lopatto
},
journal = { Biometrika },
year = { 2026 },
doi = { 10.1093/biomet/asag010 },
url = { https://doi.org/10.1093/biomet/asag010 }
}