The Structure of Ellipsoidal Distributions, II. Principal Components (Q3750820)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3988504
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | The Structure of Ellipsoidal Distributions, II. Principal Components |
scientific article; zbMATH DE number 3988504 |
Statements
The Structure of Ellipsoidal Distributions, II. Principal Components (English)
0 references
1986
0 references
Hotelling's principal components analysis
0 references
Principal-axis rotations
0 references
ellipsoidal distributions
0 references
normal-theory tests
0 references
invariance properties
0 references
sample dispersion
0 references
0.8311418294906616
0 references
0.7991637587547302
0 references
0.7943329811096191
0 references
0.7777421474456787
0 references
[For part I see ibid. 26, 779-787 (1984; Zbl 0565.62038).] Principal-axis rotations are informative for data from ellipsoidal distributions even without moments. The standard normal-theory tests used in principal components analysis are shown to be identical in level and power for all ellipsoidal distributions under a specified model for sampling. This follows from basic invariance properties of distributions associated with sample dispersion matrices. Similar findings are reported for \textit{H. Hotelling}'s [Biometrika 28, 321-377 (1936; Zbl 0015.40705)] canonical analysis in part I of this study.
0 references