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
Author
Haoge Chang
Author
Patrick Lopatto
Author
Research Topics & Keywords
Causal Inference
Research Area
Experimental Design
Research Area
Citation 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 }
}