A characterization of multivariate l1-norm symmetric distributions
Let z be an n x 1 interchangeable random vector and z(1) [less-than-or-equals, slant] ... [less-than-or-equals, slant] z(n) be its order statistics. Let ui = (n - i + 1) x (z(i) - z(i - 1))i = 1, ... n, with z(0) = 0 and U = (u1,...,Un)'. The main result is that z = u iff z is a multivariate l1-norm symmetric distribution.
Year of publication: |
1989
|
---|---|
Authors: | Fang, Bi-Qi ; Fang, Kai-Tai |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 7.1989, 4, p. 297-299
|
Publisher: |
Elsevier |
Keywords: | multivariate l1-norm symmetric distribution normalized spacing order statistics survival function |
Saved in:
Saved in favorites
Similar items by person
-
Some families of mutivariate symmetric distributions related to exponential distribution
Fang, Kai-Tai, (1988)
-
Winker, Peter, (1999)
-
A note on the mixed geographically weighted regression model
Mei, Chang-Lin, (2004)
- More ...