Identifiability of finite mixtures of elliptical distributions

被引：67
|
作者：
Holzmann, Hajo
Munk, Axel
Gneiting, Tilmann
机构：
[1] Univ Gottingen, Inst Math Stochast, D-37073 Gottingen, Germany
[2] Univ Washington, Dept Stat, Seattle, WA 98195 USA
来源：
SCANDINAVIAN JOURNAL OF STATISTICS | 2006年 / 33卷 / 04期
关键词：
characteristic function; elliptically symmetric; finite mixture; identifiability; Laplace transform; multivariate t distribution; normal scale mixture;
D O I：
10.1111/j.1467-9469.2006.00505.x
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
We present general results on the identifiability of finite mixtures of elliptical distributions under conditions on the characteristic generators or density generators. Examples include the multivariate t-distribution, symmetric stable laws, exponential power and Kotz distributions. In each case, the shape parameter is allowed to vary in the mixture, in addition to the location vector and the scatter matrix. Furthermore, we discuss the identifiability of finite mixtures of elliptical densities with generators that correspond to scale mixtures of normal distributions.
引用
收藏
页码：753 / 763
页数：11
相关论文
共 50 条
12345