Permutation tests

M. Hallin, C. Ley

    Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review

    Abstract

    Permutation tests are an interesting and conceptually simple alternative to traditional tests when the required distributional assumptions (typically, Gaussian assumptions) are likely to be violated. We start with three motivational examples explaining how to construct permutation and sampled permutation tests, and provide empirical evidence of the way they outperform their traditional competitors onseveral counts: validity, unbiasedness, and power. We then show how permutation tests follow from more traditional Neyman alpha-structure arguments, and discuss some asymptotic results establishing that extended validity, in this context, does not imply any loss of asymptotic efficiency.
    Original languageEnglish
    Title of host publicationEncyclopedia of Environmetrics, 2nd Edition
    EditorsW. Piegorsch, A. El Shaarawi
    PublisherWiley
    Pages209-210
    Number of pages3510
    ISBN (Print)9780470973882
    Publication statusPublished - 2012

    Fingerprint

    Asymptotic efficiency
    Competitors
    Empirical evidence
    Unbiasedness

    Cite this

    Hallin, M., & Ley, C. (2012). Permutation tests. In W. Piegorsch, & A. El Shaarawi (Eds.), Encyclopedia of Environmetrics, 2nd Edition (pp. 209-210). Wiley.
    Hallin, M. ; Ley, C. / Permutation tests. Encyclopedia of Environmetrics, 2nd Edition. editor / W. Piegorsch ; A. El Shaarawi. Wiley, 2012. pp. 209-210
    @inbook{8263d3623c814d05a9dee58935dc8888,
    title = "Permutation tests",
    abstract = "Permutation tests are an interesting and conceptually simple alternative to traditional tests when the required distributional assumptions (typically, Gaussian assumptions) are likely to be violated. We start with three motivational examples explaining how to construct permutation and sampled permutation tests, and provide empirical evidence of the way they outperform their traditional competitors onseveral counts: validity, unbiasedness, and power. We then show how permutation tests follow from more traditional Neyman alpha-structure arguments, and discuss some asymptotic results establishing that extended validity, in this context, does not imply any loss of asymptotic efficiency.",
    author = "M. Hallin and C. Ley",
    note = "Pagination: 3510",
    year = "2012",
    language = "English",
    isbn = "9780470973882",
    pages = "209--210",
    editor = "W. Piegorsch and {El Shaarawi}, A.",
    booktitle = "Encyclopedia of Environmetrics, 2nd Edition",
    publisher = "Wiley",

    }

    Hallin, M & Ley, C 2012, Permutation tests. in W Piegorsch & A El Shaarawi (eds), Encyclopedia of Environmetrics, 2nd Edition. Wiley, pp. 209-210.

    Permutation tests. / Hallin, M.; Ley, C.

    Encyclopedia of Environmetrics, 2nd Edition. ed. / W. Piegorsch; A. El Shaarawi. Wiley, 2012. p. 209-210.

    Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review

    TY - CHAP

    T1 - Permutation tests

    AU - Hallin, M.

    AU - Ley, C.

    N1 - Pagination: 3510

    PY - 2012

    Y1 - 2012

    N2 - Permutation tests are an interesting and conceptually simple alternative to traditional tests when the required distributional assumptions (typically, Gaussian assumptions) are likely to be violated. We start with three motivational examples explaining how to construct permutation and sampled permutation tests, and provide empirical evidence of the way they outperform their traditional competitors onseveral counts: validity, unbiasedness, and power. We then show how permutation tests follow from more traditional Neyman alpha-structure arguments, and discuss some asymptotic results establishing that extended validity, in this context, does not imply any loss of asymptotic efficiency.

    AB - Permutation tests are an interesting and conceptually simple alternative to traditional tests when the required distributional assumptions (typically, Gaussian assumptions) are likely to be violated. We start with three motivational examples explaining how to construct permutation and sampled permutation tests, and provide empirical evidence of the way they outperform their traditional competitors onseveral counts: validity, unbiasedness, and power. We then show how permutation tests follow from more traditional Neyman alpha-structure arguments, and discuss some asymptotic results establishing that extended validity, in this context, does not imply any loss of asymptotic efficiency.

    M3 - Chapter

    SN - 9780470973882

    SP - 209

    EP - 210

    BT - Encyclopedia of Environmetrics, 2nd Edition

    A2 - Piegorsch, W.

    A2 - El Shaarawi, A.

    PB - Wiley

    ER -

    Hallin M, Ley C. Permutation tests. In Piegorsch W, El Shaarawi A, editors, Encyclopedia of Environmetrics, 2nd Edition. Wiley. 2012. p. 209-210