Parallel principal axes

[摘要] Parallel principal axes are introduced and used to investigate the structure of multivariate distributions. In the class of elliptical distributions, it is shown that self-consistency of parallel principal axes characterizes normality. Parallel principal axes are used to illustrate the equivalent event fallacy in the context of self-consistency. Finally, the idea of piecewise principal component axes is introduced and related to an allometric extension model. (C) 2000 Academic Press.

[发布日期] 2000-11-01

[关键词] allometric extension;Borel's paradox;elliptical distributions;mean squared error;principal points;self-consistency;self-consistent points