Bootstrap methods for median regression models.

The least-absolute-deviations (LAD) estimator for a median-regression model does not satisfy the standard conditions for obtaining asymptotic refinements through use of the bootstrap because the LAD objective function is not smooth. This paper overcomes this problem by smoothing the objective funct...

Descripción completa

Detalles Bibliográficos
Publicado en:Econometrica Vol. 66; no. 6; pp. 1327 - 1352
Autor principal: Horowitz, Joel L.
Formato: Artículo
Publicado: Wiley-Blackwell November 1998
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=512724526&site=ehost-live
header:
  @attributes:
    shortDbName: ssf
    uiTerm: 512724526
    longDbName: Social Sciences Full Text (H.W. Wilson)
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00129682
        ECN
      jtl: Econometrica
      issn: 00129682
      maglogo: N
    pubinfo:
      dt: November 1998
      vid: 66
      iid: 6
      pid: 480
      pub: Wiley-Blackwell
    artinfo:
      ui:
        512724526
        10.2307/2999619
      ppf: 1327
      ppct: 25
      formats:
      tig:
        atl: Bootstrap methods for median regression models.
      aug:
        au: Horowitz, Joel L.
      su:
        Asymptotic expansions
        Statistical hypothesis testing
        Statistical bootstrapping
        Regression analysis
        Statistical smoothing
      sug:
        subj:
          Asymptotic expansions
          Statistical hypothesis testing
          Statistical bootstrapping
          Regression analysis
          Statistical smoothing
      ab: The least-absolute-deviations (LAD) estimator for a median-regression model does not satisfy the standard conditions for obtaining asymptotic refinements through use of the bootstrap because the LAD objective function is not smooth. This paper overcomes this problem by smoothing the objective function. The smoothed estimator is asymptotically equivalent to the standard LAD estimator. With bootstrap critical values, the rejection probabilities of symmetrical t and x2 tests based on the smoothed estimator are correct through O(n-y) under the null hypothesis, where y < 1 but can be arbitrarily close to 1. In contrast, first-order asymptotic approximations make errors of size O(n-y). These results also hold for symmetrical t and x2 tests for censored median regression models. Reprinted by permission of the Econometric Society.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: N
    holdings:
      @attributes:
        islocal: N