pcaPA: Parallel Analysis for ordinal and numeric data using polychoric
and Pearson correlations with S3 classes
A set of functions to perform parallel analysis for
principal components analysis intended mainly for large data
sets. It performs a parallel analysis of continuous, ordered
(including dichotomous/binary as a special case) or mixed type
of data associated with a principal components analysis.
Polychoric correlations among ordered variables, Pearson
correlations among continuous variables and polyserial
correlation between mixed type variables (one ordered and one
continuous) are used. Whenever the use of polyserial or
polychoric correlations yields a non positive definite
correlation matrix, the resulting matrix is transformed into
the nearest positive definite matrix.
||R (≥ 3.0.0), polycor, ltm, stats, ggplot2, mc2d
||Carlos A. Arias and Victor H. Cervantes.
||Carlos A. Arias <carias at icfes.gov.co>
||GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]