PCAVR: A portable laboratory program for performing varimax-rotated principal components analysis of event-related potentials

Abstract

A portable laboratory computer program for performing varimax-rotated principal components analysis (PCA) of event-related potentials (ERPs) is described. The program is written in FORTRAN 77; its compiled version requires 429,140 bytes of memory. The program reads a matrix of numbers from an input file. The PCA can be performed on either the variance/covariance or the correlation matrix. The program computes the first six principal components. The user is given the option of rotating as many of the principal components as desired based upon the percentages of variance that they account for. Eigenvalues, percentages of variance, cumulative percentages of variance, factor loadings, and factor scores are written to an output file. Description of the program is preceded by a conceptual overview both of PCA as a factor analytic technique and of the application of PCA to ERP data analysis.