论文信息 - On tests of independence for spherical data-invariance and centering

On tests of independence for spherical data-invariance and centering

There are a large number of tests available for testing the independence of two circular, two spherical, or two hyper-spherical responses. We first investigate the implications of invariance and various forms of the maximal invariant, and this leads to a new test statistic. Next, we introduce a model for circular dependence to compare the power of all the tests.

Grace S. Shieh | Richard A. Johnson

[1] Nicholas I. Fisher,et al. Correlation coefficients for random variables on a unit sphere or hypersphere , 1986 .

[2] R. Courant,et al. Introduction to Calculus and Analysis , 1991 .

[3] K. Mardia,et al. A general correlation coefficient for directional data and related regression problems , 1980 .

[4] Louis-Paul Rivest,et al. Some Statistical Methods for Bivariate Circular Data , 1982 .

[5] J. Mackenzie,et al. The estimation of an orientation relationship , 1957 .

[6] N. Fisher,et al. A correlation coefficient for circular data , 1983 .

[7] R. Johnson,et al. Measures and models for angular correlation and angular-linear correlation. [correlation of random variables] , 1976 .

[8] P. E. Jupp. A nonparametric correlation coefficient and a two-sample test for random vectors or directions , 1987 .

[9] Nicholas I. Fisher,et al. Nonparametric measures of angular-angular association , 1981 .