Testing multinormality based on low-dimensional projection (Q1973320): Difference between revisions
From MaRDI portal
Latest revision as of 00:01, 26 July 2024
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Testing multinormality based on low-dimensional projection |
scientific article |
Statements
Testing multinormality based on low-dimensional projection (English)
0 references
14 December 2000
0 references
The aim of this paper is to construct projection tests for multivariate normality by making use of affine invariant statistics. These statistics can be used for arbitrary dimension \(d\) even in the case when the sample size \(n\leq d\). The construction is essentially based on properties of left-spherical matrix distributions. A practical example is analyzed.
0 references
numerical example
0 references
left-spherical matrix distribution
0 references
Mardia's skewness and kurtosis
0 references
projection test
0 references
testing multinormality
0 references
affine invariant statistics
0 references
0 references
