epVari: Varimax rotation on loadings and factor scores.

In HerveAbdi/data4PCCAR: Some data sets and R-functions for PCA and CA to accompany Abdi & Beaton (to appear, 2023) Principal Component and Correspondence Analysis with R

Varimax rotation on loadings and factor scores.

Description

epVari: a Varimax rotation on loadings and factor scores. epVari is based on stats::varimax (with parameters normalize = TRUE, eps = 1e-6), but gives also the rotated factor scores, pseudo eigenvalues and explained percentage of inertia for the rotated dimensions.

Usage

epVari(resExpo, dim2keep = 2, normalize = TRUE, ...)

Arguments

resExpo	the results of a principal component analysis performed by ExPosition::epPCA.
dim2keep	(Default = 2) the number of dimensions to keep.
normalize	(default = TRUE). Passed to varimax: logical. Should Kaiser normalization be performed? If so the loadings are re-scaled to unit length before rotation, and scaled back afterwards.
...	stuff to pass to varimax.

Details

epVar is a wrapper around stats::varimax.

Value

A list with 6 elements:

1. rotationMatrix: the rotation matrix.

2. scaled.rotationMatrix: the scaled rotation matrix (differs from rotationMatrix only when normalize = TRUE), see varimax for more.

3. rotated.I: the rotated factor scores (I-set).

4. rotated.J: the loadings (J-set).

5. rotated.eigs: the rotated pseudo-eigenvalues.

6. rotated.t: the rotated percentage of explained inertia.

Author(s)

Hervé Abdi

See Also

varimax

Examples

data(iris)
library(ExPosition)
resPCA    <- ExPosition::epPCA(iris[,1:4], graphs = FALSE)
iris.Vari <-  epVari(resPCA)

HerveAbdi/data4PCCAR documentation built on Sept. 11, 2022, 4:19 p.m.