Research

Paper

TESTING February 24, 2026

Fast Algorithms for Exact Confidence Intervals in Randomized Experiments with Binary Outcomes

Authors

Peng Zhang

Abstract

We construct exact confidence intervals for the average treatment effect in randomized experiments with binary outcomes using sequences of randomization tests. Our approach does not rely on large-sample approximations and is valid for all sample sizes. Under a balanced Bernoulli design or a matched-pairs design, we show that exact confidence intervals can be computed using only $O(\log n)$ randomization tests, yielding an exponential reduction in the number of tests compared to brute-force. We further prove an information-theoretic lower bound showing that this rate is optimal. In contrast, under balanced complete randomization, the most efficient known procedures require $O(n\log n)$ randomization tests (Aronow et al., 2023), establishing a sharp separation between these designs. In addition, we extend our algorithm to general Bernoulli designs using $O(n^2)$ randomization tests.

Metadata

arXiv ID: 2602.20498
Provider: ARXIV
Primary Category: stat.ME
Published: 2026-02-24
Fetched: 2026-02-25 06:05

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.20498v1</id>\n    <title>Fast Algorithms for Exact Confidence Intervals in Randomized Experiments with Binary Outcomes</title>\n    <updated>2026-02-24T02:53:34Z</updated>\n    <link href='https://arxiv.org/abs/2602.20498v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.20498v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We construct exact confidence intervals for the average treatment effect in randomized experiments with binary outcomes using sequences of randomization tests. Our approach does not rely on large-sample approximations and is valid for all sample sizes. Under a balanced Bernoulli design or a matched-pairs design, we show that exact confidence intervals can be computed using only $O(\\log n)$ randomization tests, yielding an exponential reduction in the number of tests compared to brute-force. We further prove an information-theoretic lower bound showing that this rate is optimal. In contrast, under balanced complete randomization, the most efficient known procedures require $O(n\\log n)$ randomization tests (Aronow et al., 2023), establishing a sharp separation between these designs. In addition, we extend our algorithm to general Bernoulli designs using $O(n^2)$ randomization tests.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='stat.ME'/>\n    <published>2026-02-24T02:53:34Z</published>\n    <arxiv:primary_category term='stat.ME'/>\n    <author>\n      <name>Peng Zhang</name>\n    </author>\n  </entry>"
}