KrzProjection: Compare matrices via Modified Krzanowski Correlation

Description

Calculates the modified Krzanowski correlation between matrices, projecting the variance in each principal components of the first matrix in to the ret.dim.2 components of the second matrix.

Usage

KrzProjection(cov.x, cov.y, ...)
## S3 method for class 'default':
KrzProjection(cov.x, cov.y, ret.dim.1 = NULL,
  ret.dim.2 = NULL, ...)
## S3 method for class 'list':
KrzProjection(cov.x, cov.y = NULL, ret.dim.1 = NULL,
  ret.dim.2 = NULL, parallel = FALSE, full.results = FALSE, ...)

Arguments

cov.x

Single covariance matrix ou list of covariance matrices. If cov.x is a single matrix is suplied, it is compared to cov.y. If cov.x is a list of matrices is suplied and no cov.y is suplied, all matrices are compared between each other. If cov.x is a list o

cov.y

First argument is compared to cov.y. If cov.x is a list, every element in cov.x is projected in cov.y.

...

aditional arguments passed to other methods

ret.dim.1

number of retained dimensions for first matrix in comparison, default for nxn matrix is n/2-1

ret.dim.2

number of retained dimensions for second matrix in comparison, default for nxn matrix is n/2-1

parallel

if TRUE computations are done in parallel. Some foreach backend must be registered, like doParallel or doMC.

full.results

if FALSE returns only total variance, if TRUE also per PC variance.

Value

Ratio of projected variance to total variance, and ratio of projected total in each PC

References

Krzanowski, W. J. (1979). Between-Groups Comparison of Principal Components. Journal of the American Statistical Association, 74(367), 703. doi:10.2307/2286995

Examples

Run this code

c1 <- RandomMatrix(10)
c2 <- RandomMatrix(10)
KrzProjection(c1, c2)

m.list <- RandomMatrix(10, 3)
KrzProjection(m.list)
KrzProjection(m.list, full.results = TRUE)
KrzProjection(m.list, ret.dim.1 = 5, ret.dim.2 = 4)
KrzProjection(m.list, ret.dim.1 = 4, ret.dim.2 = 5)

KrzProjection(m.list, c1)
KrzProjection(m.list, c1, full.results = TRUE)

#Multiple threads can be used with some foreach backend library, like doMC or doParallel
#library(doParallel)
##Windows:
#cl <- makeCluster(2)
#registerDoParallel(cl)
##Mac and Linux:
#registerDoParallel(cores = 2)
#KrzProjection(m.list, parallel = TRUE)