Non-parametric estimation of copula based mutual information

被引:2
作者
Krishnankutty, Baby Alpettiyil [1 ]
Ganapathy, Rajesh [1 ]
Sankaran, Paduthol Godan [1 ]
机构
[1] Univ Sci & Technol, Dept Stat Cochin, Cochin 22, Kerala, India
关键词
mutual information; copula entropy; non-parametric estimation; kernel; DENSITY; ASSOCIATION; ENTROPY;
D O I
10.1080/03610926.2018.1563180
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
Mutual information is a measure for investigating the dependence between two random variables. The copula based estimation of mutual information reduces the complexity because it is depend only on the copula density. We propose two estimators and discuss the asymptotic properties. To compare the performance of the estimators a simulation study is carried out. The methods are illustrated using real data sets.
引用
收藏
页码:1513 / 1527
页数:15
相关论文
共 23 条
[1]   ENTROPY EXPRESSIONS AND THEIR ESTIMATORS FOR MULTIVARIATE DISTRIBUTIONS [J].
AHMED, NA ;
GOKHALE, DV .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1989, 35 (03) :688-692
[2]  
[Anonymous], 2006, Copulas: From theory to application in finance, chapter the estimation of copulas: Theory and Practice
[3]  
[Anonymous], 2007, INTRO COPULAS, DOI DOI 10.1007/0-387-28678-0
[4]  
[Anonymous], 2015, INTRO PROBABILITY ST
[5]  
[Anonymous], 1993, HDB SMALL DATA SETS
[6]  
Behnen Konrad., 1985, Statistics Risk Modeling, V3, P239
[7]  
Beirlant J, 1997, INT J MATH STAT SCI, V6, P17
[8]   Mutual information as a measure of multivariate association: analytical properties and statistical estimation [J].
Blumentritt, Thomas ;
Schmid, Friedrich .
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2012, 82 (09) :1257-1274
[9]   Nonparametric estimation of copula functions for dependence modelling [J].
Chen, Song Xi ;
Huang, Tzee-Ming .
CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, 2007, 35 (02) :265-282
[10]   Beta kernel estimators for density functions [J].
Chen, SX .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1999, 31 (02) :131-145