The asymptotic normality of linear functions of conditional order statistics (Q3787295)
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 4050767
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | The asymptotic normality of linear functions of conditional order statistics |
scientific article; zbMATH DE number 4050767 |
Statements
The asymptotic normality of linear functions of conditional order statistics (English)
0 references
1986
0 references
asymptotic normality of linear functions of conditional order statistics
0 references
conditional distribution
0 references
0 references
0.8320310115814209
0 references
0.8160297870635986
0 references
0.7980198264122009
0 references
0.7966384291648865
0 references
Let F(y\(| x)\) be the conditional distribution of \(Y\in R\) d when \(X=x\in R\) 1, \(\xi_{\alpha,x}=\sup \{y:\) \(F(y| x)<\alpha \}\), J be a measurable function on (0,1) and \(C_ i\) be some constants. To estimate \[ \theta (x)=\int^{\infty}_{-\infty}yJ(F(y| x))F(dy| x)\quad and \] \[ \theta \quad *(x)=\int^{\infty}_{- \infty}yJ(F(y| x))F(dy| x)+\sum^{\ell}_{i=1}C_ i \xi \quad_{p_ i,x} \] on the basis of i.i.d. observations \((X_ 1,Y_ 1),...,(X_ n,Y_ n)\), the estimates \[ \theta_ n(x)=k^{- 1}\sum^{k}_{i=1}J(i/(k+1))Y_{k,1}\quad and \] \[ \theta \quad *_ n(x)=k^{- 1}\sum^{k}_{i=1}J(i/(k+1))Y_{k,1}+\sum^{\ell}_{i=1\quad}C_ iY_{k,[kp_ i]+1} \] where \(Y_{k,1},...,Y_{k,k}\) are the conditional order statistics, \(k=k_ n\to \infty\), \(kn^{-1}\to 0\), are used. The asymptotic normality of \(\theta_ n\) and \(\theta\) \(*_ n\) is proved.
0 references