Daniel S. T. Hughes > Statistics-PCA-Varimax-0.0-2 > Statistics::PCA::Varimax
Module Version: 0.0.2

# NAME

Statistics::PCA::Varimax - A Perl implementation of Varimax rotation.

# VERSION

This document describes Statistics::PCA::Varimax version 0.0.2

# SYNOPSIS

```    use Statistics::PCA::Varimax;

# Each nested array ref corresponds to the loadings for a single factor.
[qw/  0.28681878905  0.69807334810  0.74438876316  0.47052419229  0.68079195447  0.49817011866  0.86049803480  0.64178962603 0.29784558460 /],
[qw/  0.07560334830  0.15335493657 -0.40959477002  0.52231277744 -0.15586396086 -0.49832262559 -0.11502014276  0.32160898539 0.59537280152 /],
[qw/ -0.84084848877 -0.08371208961  0.02047721303 -0.13507580587  0.14832508991  0.25345619152 -0.01159349490 -0.04396749541 0.53340721684 /],
];

}

print qq{\nOrthogonal Matrix:\n};
for my \$r (0..\$#{\$orthogonal_matrix_ref}) { for my \$c (0..\$#{\$orthogonal_matrix_ref->[\$r]}) {
print qq{\$orthogonal_matrix_ref->[\$r][\$c]\t} }; print qq{\n};
}```

# DESCRIPTION

Varimax rotation is a change of coordinates used in principal component analysis and factor analysis that maximizes the sum of the variances of the squared loadings matrix. This module exports a single routine 'rotate'. This routine is called in LIST context and accepts a LIST-of-LISTS (LoL) corresponding to the loadings matrix of a factor analysis and returns two references to LoLs (NOTE: each nested LIST corresponds to the loadings for a single factor). The first is a LoL of the rotated loadings and the seconds is a LoL of the orthogonal matrix. See http://en.wikipedia.org/wiki/Varimax_rotation.

# DEPENDENCIES

'Math::GSL::Linalg::SVD' => '0.0.2', 'Math::MatrixReal' => '2.05', 'List::Util' => '1.22',

# AUTHOR

Daniel S. T. Hughes `<dsth@cpan.net>`

Copyright (c) 2009, Daniel S. T. Hughes `<dsth@cantab.net>`. All rights reserved.