Promax.only is an oblique rotation of factor loadings. This function is directly derived from the
Promax function in the psych package, but only performs the promax rotation without first specifying
a varimax orthogonal rotation. Further specifying the power of the fitting function allows for
greater versatility.
Usage
Promax.only(x, m = 4, rotate.structure=NULL)
Arguments
x
matrix of factor loadings
m
power of fitting function
rotate.structure
rotation matrix if loadings have been prerotated. Default is the identity matrix.
Details
An oblique factor rotation will rescale the loadings with factors having correlated structure Phi
Value
loadings
Oblique factor loadings
rotmat
Rotation matrix structure. If rotated.structure supplied, it will be factored into rotmat.
Phi
Correlation matrix structure of Factors
Note
Adapted directly from Promax of the psych package
Author(s)
Lisa McFerrin
References
Hendrickson, A. E. and White, P. O, 1964, British Journal of Statistical Psychology, 17, 65-70.
See Also
promax, factor.pa.ginv
Examples
##compare to promax and Promax solutions
fa <- factanal( ~., 2, data = swiss)
Promax(loadings(fa))
Promax.only(loadings(fa))
Results
R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> library(HDMD)
Loading required package: psych
Loading required package: MASS
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/HDMD/Promax.only.Rd_%03d_medium.png", width=480, height=480)
> ### Name: Promax.only
> ### Title: Promax rotation (without prior Varimax rotation)
> ### Aliases: Promax.only
>
> ### ** Examples
>
> ##compare to promax and Promax solutions
> fa <- factanal( ~., 2, data = swiss)
> Promax(loadings(fa))
Call: NULL
Warning: A Heywood case was detected.
Standardized loadings (pattern matrix) based upon correlation matrix
Factor1 Factor2 h2 u2
Fertility -0.60 0.23 0.58 0.420
Agriculture -0.60 0.16 0.51 0.492
Examination 0.58 -0.36 0.73 0.270
Education 1.19 0.36 1.00 0.005
Catholic 0.33 1.15 0.94 0.061
Infant.Mortality -0.03 0.18 0.04 0.960
Factor1 Factor2
SS loadings 2.34 1.45
Proportion Var 0.39 0.24
Cumulative Var 0.39 0.63
Proportion Explained 0.62 0.38
Cumulative Proportion 0.62 1.00
Factor1 Factor2
Factor1 1.00 -0.65
Factor2 -0.65 1.00
> Promax.only(loadings(fa))
$loadings
Loadings:
Factor1 Factor2
Fertility -0.595 0.227
Agriculture -0.599 0.160
Examination 0.577 -0.360
Education 1.192 0.363
Catholic 0.326 1.147
Infant.Mortality 0.180
Factor1 Factor2
SS loadings 2.574 1.686
Proportion Var 0.429 0.281
Cumulative Var 0.429 0.710
$rotmat
[,1] [,2]
[1,] 1.2114085 0.4028901
[2,] 0.4957566 1.2453773
$Phi
[,1] [,2]
[1,] 1.0000000 -0.6452333
[2,] -0.6452333 1.0000000
attr(,"class")
[1] "HDMD"
>
>
>
>
>
>
> dev.off()
null device
1
>