Blending Bayesian and Classical Tools to Define Optimal Sample-Size-Dependent Significance Levels.

This article argues that researchers do not need to completely abandon the p-value, the best-known significance index, but should instead stop using significance levels that do not depend on sample sizes. A testing procedure is developed using a mixture of frequentist and Bayesian tools, with a sign...

Descripción completa

Detalles Bibliográficos
Publicado en:American Statistician Vol. 73; pp. 213 - 223
Autores principales: Gannon, Mark Andrew, de Bragança Pereira, Carlos Alberto, Polpo, Adriano
Formato: Artículo
Publicado: Taylor & Francis Ltd Mar2019 Supplement
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=136176755&site=ehost-live
header:
  @attributes:
    shortDbName: ssf
    uiTerm: 136176755
    longDbName: Social Sciences Full Text (H.W. Wilson)
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00031305
        STT
      jtl: American Statistician
      issn: 00031305
      maglogo: Y
    pubinfo:
      dt: Mar2019 Supplement
      vid: 73
      pid: 377
      pub: Taylor & Francis Ltd
    artinfo:
      ui:
        136176755
        10.1080/00031305.2018.1518268
      ppf: 213
      ppct: 10
      formats:
      tig:
        atl: Blending Bayesian and Classical Tools to Define Optimal Sample-Size-Dependent Significance Levels.
      aug:
        au:
          Gannon, Mark Andrew
          de Bragança Pereira, Carlos Alberto
          Polpo, Adriano
        affil:
          Institute of Mathematics and Statistics, University of São Paulo, São Paulo, Brazil
          Instituto de Matemática Aplicada, Universidade Federal de Mato Grosso do Sul, Campo Grande, Brazil
          Department of Statistics, Federal University of São Carlos
      su:
        Bayesian analysis
        Statistical significance
        Hardy-Weinberg formula
        Neyman-Pearson theorem
        Sample size (Statistics)
        Likelihood ratio tests
        Null hypothesis
      sug:
        subj:
          Bayesian analysis
          Statistical significance
          Hardy-Weinberg formula
          Neyman-Pearson theorem
          Sample size (Statistics)
          Likelihood ratio tests
          Null hypothesis
      keyword:
        Hardy–Weinberg equilibrium
        Neyman–Pearson lemma
        Predictive distribution
        Significance test
        Hardy–Weinberg equilibrium
        Neyman–Pearson lemma
        Predictive distribution
        Significance test
      ab: This article argues that researchers do not need to completely abandon the p-value, the best-known significance index, but should instead stop using significance levels that do not depend on sample sizes. A testing procedure is developed using a mixture of frequentist and Bayesian tools, with a significance level that is a function of sample size, obtained from a generalized form of the Neyman–Pearson Lemma that minimizes a linear combination of α, the probability of rejecting a true null hypothesis, and β, the probability of failing to reject a false null, instead of fixing α and minimizing β. The resulting hypothesis tests do not violate the Likelihood Principle and do not require any constraints on the dimensionalities of the sample space and parameter space. The procedure includes an ordering of the entire sample space and uses predictive probability (density) functions, allowing for testing of both simple and compound hypotheses. Accessible examples are presented to highlight specific characteristics of the new tests.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: N
    holdings:
      @attributes:
        islocal: N