Главная

Pierre Neuvial

  1. On false discovery rate thresholding for classification under sparsity.

    Пт, 07/01/2011 - 15:01 — SomNambul
    Authors: Etienne Roquain, Pierre Neuvial
    Subjects: Methodology
    Abstract

    We study the properties of false discovery rate (FDR) thresholding, viewed as
    a classification procedure. The "0"-class (null) is assumed to have a known,
    symmetric log-concave density while the "1"-class (alternative) is obtained
    from the "0"-class either by translation (location model) or by scaling (scale
    model). Furthermore, the "1"-class is assumed to have a small number of
    elements w.r.t. the "0"-class (sparsity). Non-asymptotic oracle inequalities
    are derived for the excess risk of FDR thresholding.

  2. Intrinsic Bounds and False Discovery Rate Control in Multiple Testing Problems.

    Чт, 03/04/2010 - 16:01 — SomNambul
    Authors: Pierre Neuvial
    Subjects: Statistics
    Abstract

    When testing a large number of independent hypotheses, three different
    questions are of interest: are some hypotheses true alternatives? How many of
    them? Which of them? These questions give rise to a detection, an estimation,
    and a selection problem. Recent work demonstrates the existence of intrinsic
    bounds in these problems: detection and estimation boundaries in sparse
    location models, and criticality for the selection problem. We study
    consequences of such limitations in terms of power of False Discovery Rate
    (FDR) controlling procedures.

Math-arch

RSS-материал