On R-package: seeded CCA

Classical canonical correlation analysis is one of popular dimension reduction methodologies to reduce the dimensions of two sets of variables. However, the methodology is limited in use for so-called large p small n data, because matrix inversions are not possible in such data. So-called seeded canonical correlation analysis is recently developed to overcome this limitation. The package of seedCCA is originated to practically implement the seeded canonical correlation analysis in R. The statistical language R is a license-free statistical software, so it is popular to most academic and industrial statisticians. The main goal of the paper is to introduce seed-CCA and to present two real data examples to show how to use the package. It is expected that the package along with the paper will contribute to high-dimensional data analysis.