Permutation tests for general dependent truncation

Alzheimer's Disease Neuroimaging Initiative, Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging, Harvard Aging Brain Study

Research output: Contribution to journalArticle

Abstract

Truncated survival data arise when the event time is observed only if it falls within a subject-specific region, known as the truncation set. Left-truncated data arise when there is delayed entry into a study, such that subjects are included only if their event time exceeds some other time. Quasi-independence of truncation and failure refers to factorization of their joint density in the observable region. Under quasi-independence, standard methods for survival data such as the Kaplan–Meier estimator and Cox regression can be applied after simple adjustments to the risk sets. Unlike the requisite assumption of independent censoring, quasi-independence can be tested, e.g., using a conditional Kendall's tau test. Current methods for testing for quasi-independence are powerful for monotone alternatives. Nonetheless, it is essential to detect any kind of deviation from quasi-independence so as not to report a biased Kaplan–Meier estimator or regression effect, which would arise from applying the simple risk set adjustment when dependence holds. Nonparametric, minimum p-value tests that are powerful against non-monotone alternatives are developed to offer protection against erroneous assumptions of quasi-independence. The use of conditional and unconditional methods of permutation for evaluation of the proposed tests is investigated in simulation studies. The proposed tests are applied to a study on the cognitive and functional decline in aging.

Original languageEnglish (US)
Pages (from-to)308-324
Number of pages17
JournalComputational Statistics and Data Analysis
Volume128
DOIs
StatePublished - Dec 1 2018

Fingerprint

Quasi-independence
Permutation Test
Truncation
Dependent
Factorization
Kaplan-Meier Estimator
Truncated Data
Aging of materials
Survival Data
Adjustment
Testing
Regression Effects
Cox Regression
Kendall's tau
Alternatives
p-Value
Censoring
Biased
Monotone
Exceed

Keywords

  • Kendall's tau
  • Minimally selected test
  • Monotone dependence
  • Quasi-independence

ASJC Scopus subject areas

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Alzheimer's Disease Neuroimaging Initiative, Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging, & Harvard Aging Brain Study (2018). Permutation tests for general dependent truncation. Computational Statistics and Data Analysis, 128, 308-324. https://doi.org/10.1016/j.csda.2018.07.012

Permutation tests for general dependent truncation. / Alzheimer's Disease Neuroimaging Initiative; Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging; Harvard Aging Brain Study.

In: Computational Statistics and Data Analysis, Vol. 128, 01.12.2018, p. 308-324.

Research output: Contribution to journalArticle

Alzheimer's Disease Neuroimaging Initiative, Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging & Harvard Aging Brain Study 2018, 'Permutation tests for general dependent truncation', Computational Statistics and Data Analysis, vol. 128, pp. 308-324. https://doi.org/10.1016/j.csda.2018.07.012
Alzheimer's Disease Neuroimaging Initiative, Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging, Harvard Aging Brain Study. Permutation tests for general dependent truncation. Computational Statistics and Data Analysis. 2018 Dec 1;128:308-324. https://doi.org/10.1016/j.csda.2018.07.012
Alzheimer's Disease Neuroimaging Initiative ; Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging ; Harvard Aging Brain Study. / Permutation tests for general dependent truncation. In: Computational Statistics and Data Analysis. 2018 ; Vol. 128. pp. 308-324.
@article{cbeb1149408b4cca9be9d664560e5e35,
title = "Permutation tests for general dependent truncation",
abstract = "Truncated survival data arise when the event time is observed only if it falls within a subject-specific region, known as the truncation set. Left-truncated data arise when there is delayed entry into a study, such that subjects are included only if their event time exceeds some other time. Quasi-independence of truncation and failure refers to factorization of their joint density in the observable region. Under quasi-independence, standard methods for survival data such as the Kaplan–Meier estimator and Cox regression can be applied after simple adjustments to the risk sets. Unlike the requisite assumption of independent censoring, quasi-independence can be tested, e.g., using a conditional Kendall's tau test. Current methods for testing for quasi-independence are powerful for monotone alternatives. Nonetheless, it is essential to detect any kind of deviation from quasi-independence so as not to report a biased Kaplan–Meier estimator or regression effect, which would arise from applying the simple risk set adjustment when dependence holds. Nonparametric, minimum p-value tests that are powerful against non-monotone alternatives are developed to offer protection against erroneous assumptions of quasi-independence. The use of conditional and unconditional methods of permutation for evaluation of the proposed tests is investigated in simulation studies. The proposed tests are applied to a study on the cognitive and functional decline in aging.",
keywords = "Kendall's tau, Minimally selected test, Monotone dependence, Quasi-independence",
author = "{Alzheimer's Disease Neuroimaging Initiative} and {Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging} and {Harvard Aging Brain Study} and Chiou, {Sy Han} and Jing Qian and Elizabeth Mormino and Rebecca Betensky",
year = "2018",
month = "12",
day = "1",
doi = "10.1016/j.csda.2018.07.012",
language = "English (US)",
volume = "128",
pages = "308--324",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",

}

TY - JOUR

T1 - Permutation tests for general dependent truncation

AU - Alzheimer's Disease Neuroimaging Initiative

AU - Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging

AU - Harvard Aging Brain Study

AU - Chiou, Sy Han

AU - Qian, Jing

AU - Mormino, Elizabeth

AU - Betensky, Rebecca

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Truncated survival data arise when the event time is observed only if it falls within a subject-specific region, known as the truncation set. Left-truncated data arise when there is delayed entry into a study, such that subjects are included only if their event time exceeds some other time. Quasi-independence of truncation and failure refers to factorization of their joint density in the observable region. Under quasi-independence, standard methods for survival data such as the Kaplan–Meier estimator and Cox regression can be applied after simple adjustments to the risk sets. Unlike the requisite assumption of independent censoring, quasi-independence can be tested, e.g., using a conditional Kendall's tau test. Current methods for testing for quasi-independence are powerful for monotone alternatives. Nonetheless, it is essential to detect any kind of deviation from quasi-independence so as not to report a biased Kaplan–Meier estimator or regression effect, which would arise from applying the simple risk set adjustment when dependence holds. Nonparametric, minimum p-value tests that are powerful against non-monotone alternatives are developed to offer protection against erroneous assumptions of quasi-independence. The use of conditional and unconditional methods of permutation for evaluation of the proposed tests is investigated in simulation studies. The proposed tests are applied to a study on the cognitive and functional decline in aging.

AB - Truncated survival data arise when the event time is observed only if it falls within a subject-specific region, known as the truncation set. Left-truncated data arise when there is delayed entry into a study, such that subjects are included only if their event time exceeds some other time. Quasi-independence of truncation and failure refers to factorization of their joint density in the observable region. Under quasi-independence, standard methods for survival data such as the Kaplan–Meier estimator and Cox regression can be applied after simple adjustments to the risk sets. Unlike the requisite assumption of independent censoring, quasi-independence can be tested, e.g., using a conditional Kendall's tau test. Current methods for testing for quasi-independence are powerful for monotone alternatives. Nonetheless, it is essential to detect any kind of deviation from quasi-independence so as not to report a biased Kaplan–Meier estimator or regression effect, which would arise from applying the simple risk set adjustment when dependence holds. Nonparametric, minimum p-value tests that are powerful against non-monotone alternatives are developed to offer protection against erroneous assumptions of quasi-independence. The use of conditional and unconditional methods of permutation for evaluation of the proposed tests is investigated in simulation studies. The proposed tests are applied to a study on the cognitive and functional decline in aging.

KW - Kendall's tau

KW - Minimally selected test

KW - Monotone dependence

KW - Quasi-independence

UR - http://www.scopus.com/inward/record.url?scp=85051117982&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051117982&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2018.07.012

DO - 10.1016/j.csda.2018.07.012

M3 - Article

VL - 128

SP - 308

EP - 324

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

ER -