Optimum stratification for allocation proportional to strata totals for simple random sampling scheme (Q3678452)

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 3899926
Language Label Description Also known as
default for all languages
No label defined
    English
    Optimum stratification for allocation proportional to strata totals for simple random sampling scheme
    scientific article; zbMATH DE number 3899926

      Statements

      Optimum stratification for allocation proportional to strata totals for simple random sampling scheme (English)
      0 references
      0 references
      0 references
      1984
      0 references
      minimal equations
      0 references
      limiting lower bound
      0 references
      series expansions
      0 references
      optimum boundary problem
      0 references
      stratified sampling
      0 references
      stratification
      0 references
      auxiliary variable
      0 references
      superpopulation approach
      0 references
      distribution-free boundary formulas
      0 references
      The paper contains a purely theoretical treatment of the optimum boundary problem in stratified sampling. It is based on the more realistic assumption that the stratification ought to be based on an auxiliary variable. With the aid of the superpopulation approach distribution-free boundary formulas are presented. The boundaries need to be computed iteratively as the underlying equations are implicit. It should be noted that the boundaries resemble in a sense those used for cum \({}^ 3\sqrt{f}\) stratification. The authors display a limiting property of the variance of the population mean estimator for growing number of strata L, but this theorem seems to be not very well formulated, as it is rather peculiar to assume the sample size n constant for an infinitely increasing number of strata. That is why the statement that the variance of the population mean estimator for \(L\to \infty\) has a nonzero lower limit is doubtful and it is even wrong provided auxiliary variable with zero mean.
      0 references

      Identifiers