No label defined (Q3816772)

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 4088616
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 4088616

      Statements

      1987
      0 references
      Berry-Esseen bounds
      0 references
      finite-population U-statistic
      0 references
      standard normal distribution
      0 references
      0 references
      0 references
      Let \({\mathcal A}_ N\) be a population of N balls bearing numbers \(a_{N1},...,a_{NN}\), respectively. Draw n balls from \({\mathcal A}_ N\) randomly without replacement, and denote the numbers appearing on these n balls by \(X_ 1,...,X_ n\). Suppose that \(\phi_ N(x,y)\) is a Borel- measurable function, symmetric in x and y. Let \[ U_ n=\left( \begin{matrix} n\\ 2\end{matrix} \right)^{-1}\sum_{t\leq j<k\leq n}\phi_ N(X_ j,X_ k), \] \(\theta_ N=E \phi_ N(X_ 1,X_ 2)\), \(g(X_ 1)=E\{\phi_ N(X_ 1,X_ 2)| X_ 1\}\), \(\sigma^ 2_ g=Var\{g(X_ 1)\}\). We prove that if there exist fixed constants \(\lambda_ 1\) and \(\lambda_ 2\) such that \(0<\lambda_ 1\leq n/N\triangleq p_ N\leq \lambda_ 2<1\), then \[ \sup_{x}| P(\frac{\sqrt{n}(U_ n-\theta_ N)}{2\sqrt{1- p_ N}\sigma_ g}\leq x)-\Phi (x)| \leq \frac{C}{\sqrt{n}}\{(E \phi^ 2_ N(X_ 1,X_ 2)/\sigma^ 2_ g)+(E| g(X_ 1)|^ 3/\sigma^ 3_ g)\}, \] where \(\Phi\) (x) is the standard normal distribution function, and C is an absolute constant depending solely upon \(\lambda_ 1\) and \(\lambda_ 2\).
      0 references

      Identifiers