No label defined (Q3986299)
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 28608
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 28608 |
Statements
27 June 1992
0 references
uniform spacings
0 references
empirical distribution function
0 references
Bahadur theorem for spacings
0 references
0.8504864573478699
0 references
0.8475099802017212
0 references
0.8463175296783447
0 references
0.845270574092865
0 references
0.8441601991653442
0 references
Let \(0 < U_{1,n} < U_{2,n} < \dots < U_{n,n} < 1\) be the ordered sample of a sample \(U_ 1,U_ 2,\dots,U_ n\) of independent uniform (0,1) r.v.'s. Consider the uniform spacings \(D_{i,n}=U_{i,n-1}- U_{i-1,n-1}\) and the corresponding empirical distribution function \(F_ n(t)=n^{-1}\sum^ n_{i=1}I_{(-\infty,t]}(nD_{i,n})\). Define the empirical process \(\delta_ n(t)=n^{1/2}(F_ n(t)-F(t))\), where \(F(t)=1-e^{-t}\) \((t\geq 0)\) is the limit distribution of \(nD_{1,n}\). Finally, let \(a_ n(s)=n^{1/2}(U_ n(s)-s)\) and \(b_ n(s)=n^{1/2}(V_ n(s)-s)\), where \(U_ n(s)=F_ n(G(s))\), \(V_ n(s)=F(G_ n(s))\) and \(G(s)\) resp. \(G_ n(s)\) are the inverse functions of \(F(s)\) resp. \(F_ n(s)\). The authors investigate the properties of \(a_ n(s)+b_ n(s)\). One of their results claims that \[ n^{1/4}(\log n)^{-1/2}\sup_ s | a_ n(s)+b_ n(s)| (\sup_ s| b_ n(s)|)^{-1}\to 1 \hbox{ in probability}. \] It is a natural analogue of the classical Bahadur theorem for spacings.
0 references