standardCCA: Internal standard CCA function.

In mixedCCA: Sparse Canonical Correlation Analysis for High-Dimensional Mixed Data

Internal standard CCA function.

Description

This function is modified from original CCA function for two reasons: to deal with only positive eigenvalues larger than the tolerance when calculating the inverse of the matrices and to compute Singular Value Decomposition using irlba algorithm. Inputs should be correlation or covariance matrices of each data set and between datasets. This function returns only the first pair of canonical covariates.

Usage

standardCCA(S1, S2, S12, tol = 1e-04)

Arguments

S1	correlation/covariance matrix of dataset X1.
S2	correlation/covariance matrix of dataset X2.
S12	correlation/covariance matrix between dataset X1 and dataset X2.
tol	tolerance for eigenvalues. standardCCA function only deals with positive eigenvalues larger than the tolerance.

Value

standardCCA returns a data.frame containing

• cancor: estimated canonical correlation.

• w1: estimated canonical direction w1.

• w2: estimated canonical direction w2.

mixedCCA documentation built on Sept. 10, 2022, 1:06 a.m.