A family of minimum quantile distance estimators for the three-parameter Weibull distribution (Q3217434)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3883401
Language Label Description Also known as
default for all languages
No label defined
    English
    A family of minimum quantile distance estimators for the three-parameter Weibull distribution
    scientific article; zbMATH DE number 3883401

      Statements

      1984
      0 references
      optimal quantile selection
      0 references
      order statistics
      0 references
      asymptotic normality
      0 references
      consistency
      0 references
      minimum quantile distance estimators
      0 references
      sample quantiles
      0 references
      3- parameter Weibull distribution
      0 references
      A family of minimum quantile distance estimators for the three-parameter Weibull distribution (English)
      0 references
      The main aim of this paper is to propose a new class of so-called minimum quantile distance estimators based on a subset of the sample quantiles and to establish the basic properties of them. More precisely, let \(X_{(1)},...,X_{(n)}\) be an ordered sample from a 3-parameter Weibull distribution. Let Q(u,\(\theta)\), \(\theta =(\mu,\sigma,c)\) being the parameter, denote the corresponding quantile function, resp. \(\hat Q(\)u) the corresponding sample quantile function. Define for a given set of \(k<n\) percentile points \(U=\{u_ 1,...,u_ k\}\), satisfying \(0<u_ 1<...<u_ k<1\), the vectors \(\hat Q_ U=(\hat Q(u_ 1),...,\hat Q(u_ k))'\), resp. \(Q_ U(\theta)=(Q(u_ 1,\theta),....,Q(u_ k,\theta))'\). Then the minimum quantile distance estimator is defined as any vector that minimizes the quadratic form \[ E(\theta)=(\hat Q_ U-Q_ U(\theta))'W(\theta)(\hat Q_ U-Q_ U(\theta)) \] as a function of \(\theta\), where W(\(\theta)\) is a user- defined \(k\times k\) matrix of weights that may be chosen to depend on \(\theta\). Authors give conditions under which the proposed estimator is consistent and asymptotically normal and discuss in details the optimal choice of vector U and matrix W. The paper is numerically illustrated.
      0 references
      0 references
      0 references
      0 references

      Identifiers