Symmetry tests for manifold-valued random variables

被引:0
作者
Jesus Salamanca, Juan [1 ]
机构
[1] Univ Oviedo, Dept Estadist & IO & DM, Escuela Politecn Ingn, E-33071 Gijon, Spain
关键词
Symmetry of a random variable; testing symmetry; manifold-valued random variable; Killing vector field; FRECHET MEANS; REGRESSION; STATISTICS;
D O I
10.1080/03610926.2019.1628990
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 start with a random variable defined on a complex metric structure. We are interested in obtaining simple tests to determine whether such random variable has a symmetry. More precisely, we consider a Riemannian manifold which admits a symmetry represented by a Killing vector field. In this setting, we take into account a random variable. Our problem is to analyze and test whether such random variable shares the symmetry related to the Killing vector field. In other words, if that random variable possesses such that symmetry. As particular cases, we consider the Euclidean space and the round sphere.
引用
收藏
页码:61 / 72
页数:12
相关论文
共 24 条
[21]   Intrinsic statistics on riemannian manifolds: Basic tools for geometric measurements [J].
Pennec, Xavier .
JOURNAL OF MATHEMATICAL IMAGING AND VISION, 2006, 25 (01) :127-154
[22]   Bernstein-type Theorems in a Riemannian Manifold with an Irrotational Killing Vector Field [J].
Romero, Alfonso ;
Rubio, Rafael M. .
MEDITERRANEAN JOURNAL OF MATHEMATICS, 2016, 13 (03) :1285-1290
[23]  
Vinue Visus G., 2014, K MEANS ALGORITHM 3D
[24]   BAYESIAN MANIFOLD REGRESSION [J].
Yang, Yun ;
Dunson, David B. .
ANNALS OF STATISTICS, 2016, 44 (02) :876-905