A Survey of Rank Tests for Classes of Tail Alternatives

Willem Albers, Cecile M. Schut

    Research output: Chapter in Book/Report/Conference proceedingChapterAcademic

    Abstract

    Quite often in the two-sample problem, the assumption of constant shift is not realistic. We study increasing shift functions, leading to tail alternatives. Several classes of such functions, most of which have been used implicitly in previous work, are compared. We investigate the robustness of the corresponding tests against misspecification of the shift function. Some attention is also devoted to the possibility of adaptation.
    Original languageEnglish
    Title of host publicationResearch developments in probability and statistics
    Subtitle of host publicationFestschrift in honor of Madan L. Puri on the occasion of his 65th birthday
    EditorsEdgar Brunner, Manfred Denker
    Place of PublicationUtrecht
    PublisherVSP
    Pages289-302
    Number of pages14
    ISBN (Print)9789067642095
    Publication statusPublished - 1996

    Keywords

    • Rank tests
    • Two-sample problem
    • Increasing shift functions
    • Tail alternatives
    • Robustness
    • Misspecification
    • Adaptation

    Fingerprint Dive into the research topics of 'A Survey of Rank Tests for Classes of Tail Alternatives'. Together they form a unique fingerprint.

    Cite this