Restricted domains of attraction of exp(-e-x)
For independent and identically distributed random variables the domain of attraction of exp(-e-x) for the maximum is investigated under the restriction that the population distribution has a density. Necessary and sufficient conditions are established in term of the expected residual life and hazard rate. Furthermore, it is shown that, for ultimately concave distributions with increasing hazard rate, the von Mises condition is both necessary and sufficient for a population distribution to be in the domain of attraction of exp(-e-x).
Year of publication: |
1987
|
---|---|
Authors: | Galambos, J. ; Obretenov, A. |
Published in: |
Stochastic Processes and their Applications. - Elsevier, ISSN 0304-4149. - Vol. 25.1987, p. 265-271
|
Publisher: |
Elsevier |
Keywords: | independent and identically distributed random variables density maximum domain of attraction of exp(-e-x) expected residual life hazard rate monotonic hazard rate necessary and sufficient condition von Mises condition |
Saved in:
Saved in favorites
Similar items by person
-
On the dependence function of Sibuya in multivariate extreme value theory
Obretenov, A., (1991)
-
Galambos, J., (1990)
-
Studies in Applied Probality: Papers in Honour of Lajos Takacs
Galambos, J., (1995)
- More ...