Limit theorems for lower-upper extreme values from two-dimensional distribution function (Q910120)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Limit theorems for lower-upper extreme values from two-dimensional distribution function |
scientific article |
Statements
Limit theorems for lower-upper extreme values from two-dimensional distribution function (English)
0 references
1990
0 references
Consider n i.i.d. two-dimensional random vectors \(\bar X_ j=(X_{1j},X_{2j})\), \(j=1,2,...,n\geq 1\). Define the order statistics of the \(t^{th}\) component by \(X_{t,1:n}\leq X_{t,2:n}\leq...\leq X_{t,n:n}\), \(t=1,2\). Let \((a_{1n},a_{2n})>0\) and \((b_{1n},b_{2n})\) be sequences of constant vectors. The limit distribution of \[ V_{k,j:n}=(a^{-1}_{1n}(X_{1,k:n}- b_{1n}),\quad a^{-1}_{2n}(X_{2,j:n}-b_{2n})),\quad 1\leq k,\quad j\leq n, \] has been obtained for properly chosen normalizing sequences, assuming the existence of the marginal limit distributions. Conditions have been developed for the asymptotic independence of the components of \(V_{k,j:n}\). Furthermore, the above results have been extended to the case when the sample size is a random variable.
0 references
weak convergence
0 references
sample of random size
0 references
bivariate extreme order statistics
0 references
transfer theory
0 references
limit distribution
0 references
normalizing sequences
0 references
marginal limit distributions
0 references
asymptotic independence
0 references