Abstract

A construction of p-values for hypothesis tests based on subsampling and the related m out of n bootstrap is introduced. The p-values are based on a modification of the usual subsampling hypothesis tests that involves an appropriate centering of the subsampled or bootstrapped test statistics as in the construction of confidence intervals. This modification makes the hypothesis tests more powerful, and as a consequence provides meaningful p-values. The procedure is shown to apply for both i.i.d. and stationary data satisfying mixing conditions. The new p-values have the desired properties of being asymptotically uniform under the null hypothesis and converging to zero under the alternative.

Original languageEnglish (US)
Pages (from-to)1358-1364
Number of pages7
JournalStatistics and Probability Letters
Volume80
Issue number17-18
DOIs
StatePublished - Sep 1 2010

Fingerprint

Subsampling
p-Value
Hypothesis Test
Mixing Conditions
Null hypothesis
Bootstrap
Test Statistic
Confidence interval
P value
Alternatives
Zero
Hypothesis test

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Berg, A., McMurry, T. L., & Politis, D. N. (2010). Subsampling p-values. Statistics and Probability Letters, 80(17-18), 1358-1364. https://doi.org/10.1016/j.spl.2010.04.018
Berg, Arthur ; McMurry, Timothy L. ; Politis, Dimitris N. / Subsampling p-values. In: Statistics and Probability Letters. 2010 ; Vol. 80, No. 17-18. pp. 1358-1364.
@article{0a84ce5d73a647eb8f6c09a6d3d9d1f5,
title = "Subsampling p-values",
abstract = "A construction of p-values for hypothesis tests based on subsampling and the related m out of n bootstrap is introduced. The p-values are based on a modification of the usual subsampling hypothesis tests that involves an appropriate centering of the subsampled or bootstrapped test statistics as in the construction of confidence intervals. This modification makes the hypothesis tests more powerful, and as a consequence provides meaningful p-values. The procedure is shown to apply for both i.i.d. and stationary data satisfying mixing conditions. The new p-values have the desired properties of being asymptotically uniform under the null hypothesis and converging to zero under the alternative.",
author = "Arthur Berg and McMurry, {Timothy L.} and Politis, {Dimitris N.}",
year = "2010",
month = "9",
day = "1",
doi = "10.1016/j.spl.2010.04.018",
language = "English (US)",
volume = "80",
pages = "1358--1364",
journal = "Statistics and Probability Letters",
issn = "0167-7152",
publisher = "Elsevier",
number = "17-18",

}

Berg, A, McMurry, TL & Politis, DN 2010, 'Subsampling p-values', Statistics and Probability Letters, vol. 80, no. 17-18, pp. 1358-1364. https://doi.org/10.1016/j.spl.2010.04.018

Subsampling p-values. / Berg, Arthur; McMurry, Timothy L.; Politis, Dimitris N.

In: Statistics and Probability Letters, Vol. 80, No. 17-18, 01.09.2010, p. 1358-1364.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Subsampling p-values

AU - Berg, Arthur

AU - McMurry, Timothy L.

AU - Politis, Dimitris N.

PY - 2010/9/1

Y1 - 2010/9/1

N2 - A construction of p-values for hypothesis tests based on subsampling and the related m out of n bootstrap is introduced. The p-values are based on a modification of the usual subsampling hypothesis tests that involves an appropriate centering of the subsampled or bootstrapped test statistics as in the construction of confidence intervals. This modification makes the hypothesis tests more powerful, and as a consequence provides meaningful p-values. The procedure is shown to apply for both i.i.d. and stationary data satisfying mixing conditions. The new p-values have the desired properties of being asymptotically uniform under the null hypothesis and converging to zero under the alternative.

AB - A construction of p-values for hypothesis tests based on subsampling and the related m out of n bootstrap is introduced. The p-values are based on a modification of the usual subsampling hypothesis tests that involves an appropriate centering of the subsampled or bootstrapped test statistics as in the construction of confidence intervals. This modification makes the hypothesis tests more powerful, and as a consequence provides meaningful p-values. The procedure is shown to apply for both i.i.d. and stationary data satisfying mixing conditions. The new p-values have the desired properties of being asymptotically uniform under the null hypothesis and converging to zero under the alternative.

UR - http://www.scopus.com/inward/record.url?scp=77954543116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954543116&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2010.04.018

DO - 10.1016/j.spl.2010.04.018

M3 - Article

AN - SCOPUS:77954543116

VL - 80

SP - 1358

EP - 1364

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

SN - 0167-7152

IS - 17-18

ER -