Modified measures of multivariate association and linear dependence
Cui Guo-sheng
Abstract
Cui Guo-sheng
Abstract
The canoical correlation coefficient of multiple variables vector is defined in this paper,and the general correlation coefficient is introduced.Moreover,it has been completely resolved that the relationships between the number of nit canonical correlation coefficient and the dimensions of linear dependence between two random vectors and the corresponding proof are given.Then that the general correlation coefficient is a generalization of correlation coefficient with one dimension random variable.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The canoical correlation coefficient of multiple variables vector is defined in this paper,and the general correlation coefficient is introduced.Moreover,it has been completely resolved that the relationships between the number of nit canonical correlation coefficient and the dimensions of linear dependence between two random vectors and the corresponding proof are given.Then that the general correlation coefficient is a generalization of correlation coefficient with one dimension random variable.
Key concepts: Distance correlation, Correlation coefficient, Fisher transformation, Mathematics, Canonical correlation, Correlation, Statistics, Multivariate random variable