Bounds on expectations of order statistics via extremal dependences
Using the concept of r-extremal dependence, which generalizes Lai and Robbins (1976) maximal dependence, we propose alternative proofs and some new results concerning expectations of order statistics with any rank, from possibly dependent variates. In particular, new, distribution-free and tight bounds are given for the expectations of order statistics from i.d. variates whose common distribution is symmetrical. Sharp approximations of the tight bounds are also given for the standard normal distribution.
Year of publication: |
1992
|
---|---|
Authors: | Gascuel, O. ; Caraux, G. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 15.1992, 2, p. 143-148
|
Publisher: |
Elsevier |
Keywords: | Expectation of order statistics extremal dependence upper and lower bounds |
Saved in:
Saved in favorites
Similar items by person
-
Bounds on distribution functions of order statistics for dependent variates
Caraux, G., (1992)
-
Approximations et majorations optimales des statistiques d'ordre d'un échantillon
Caraux, G., (1991)
-
Le développement, pour la masse ou pour une élite ?
Mainie, Ph., (1974)
- More ...