Exact anytime-valid confidence intervals for contingency tables and beyond

Rosanne J. Turner; Peter D. Grünwald

doi:10.1016/j.spl.2023.109835

Exact anytime-valid confidence intervals for contingency tables and beyond

Rosanne J. Turner^*
, Peter D. Grünwald

^*Corresponding author for this work

Psychiatry

Research output: Contribution to journal › Article › Academic › peer-review

3 Downloads (Pure)

Abstract

E-variables are tools for retaining type-I error guarantee with optional stopping. We extend E-variables for sequential two-sample tests to general null hypotheses and anytime-valid confidence sequences. We provide implementations for estimating risk difference, relative risk and odds-ratios in contingency tables.

Original language	English
Article number	109835
Journal	Statistics and Probability Letters
Volume	198
DOIs	https://doi.org/10.1016/j.spl.2023.109835
Publication status	Published - Jul 2023

Keywords

Anytime-valid
Confidence sequences
Contingency tables
Effect size

Access to Document

10.1016/j.spl.2023.109835Licence: CC BY

1-s2.0-S0167715223000597-mainFinal published version, 669 KBLicence: CC BY

Cite this

@article{39420cdf54614a76a12845c9010ef361,

title = "Exact anytime-valid confidence intervals for contingency tables and beyond",

abstract = "E-variables are tools for retaining type-I error guarantee with optional stopping. We extend E-variables for sequential two-sample tests to general null hypotheses and anytime-valid confidence sequences. We provide implementations for estimating risk difference, relative risk and odds-ratios in contingency tables.",

keywords = "Anytime-valid, Confidence sequences, Contingency tables, Effect size",

author = "Turner, \{Rosanne J.\} and Gr{\"u}nwald, \{Peter D.\}",

note = "Publisher Copyright: {\textcopyright} 2023 The Author(s)",

year = "2023",

month = jul,

doi = "10.1016/j.spl.2023.109835",

language = "English",

volume = "198",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

}

TY - JOUR

T1 - Exact anytime-valid confidence intervals for contingency tables and beyond

AU - Turner, Rosanne J.

AU - Grünwald, Peter D.

PY - 2023/7

Y1 - 2023/7

N2 - E-variables are tools for retaining type-I error guarantee with optional stopping. We extend E-variables for sequential two-sample tests to general null hypotheses and anytime-valid confidence sequences. We provide implementations for estimating risk difference, relative risk and odds-ratios in contingency tables.

AB - E-variables are tools for retaining type-I error guarantee with optional stopping. We extend E-variables for sequential two-sample tests to general null hypotheses and anytime-valid confidence sequences. We provide implementations for estimating risk difference, relative risk and odds-ratios in contingency tables.

KW - Anytime-valid

KW - Confidence sequences

KW - Contingency tables

KW - Effect size

UR - https://www.scopus.com/pages/publications/85151408465

U2 - 10.1016/j.spl.2023.109835

DO - 10.1016/j.spl.2023.109835

M3 - Article

AN - SCOPUS:85151408465

SN - 0167-7152

VL - 198

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

M1 - 109835

ER -

Exact anytime-valid confidence intervals for contingency tables and beyond

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this