Covariance matrices associated to general moments of a random vector

被引:1
|
作者
Lv, Songjun [1 ]
机构
[1] Chongqing Normal Univ, Coll Math Sci, Chongqing 401331, Peoples R China
关键词
Covariance matrix; Gaussian gauge; Power function distribution; Logistic distribution; Characterization; ENTROPY INEQUALITIES; FISHER INFORMATION; RENYI ENTROPY; SOBOLEV;
D O I
10.1016/j.jmva.2014.10.007
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
It turns out that there exist general covariance matrices associated not only to a random vector itself but also to its general moments. In this paper we introduce and characterize general covariance matrices of a random vector that are associated to some important general moments, which are determined by a specific class of convex functions. As special cases, the original covariance matrices of a random vector, as well as the pth covariance matrices characterized recently, are included. The covariance matrices associated to the p-power function distribution and the logistic distribution are characterized as by-products. (C) 2014 Elsevier Inc. All rights reserved.
引用
收藏
页码:61 / 70
页数:10
相关论文
共 50 条