pca_gca: PCA-GCA method for selecting the number of common and... In RegularizedSCA: Regularized Simultaneous Component Based Data Integration

Description

Use PCA-GCA method to identify the number of common and distinctive components.

Usage

 1 pca_gca(DATA, Jk, cor_min, return_scores)

Arguments

 DATA A concatenated data matrix with the same number of rows. Jk A vector containing number of variables in the concatinated data matrix. Please see the example below. cor_min The minimum correlation between two components. The default value is .7; thus, it means that if the correlation between the two component is at least .7, then these two components are regarded as forming a single common component. return_scores If TRUE, then the function will return the component scores for each block for further analysis.

Value

It prints out the number of components of each block and the number of common components. It also returns the component scores for each block for further analysis, if return_scores = TRUE.

Note

Please be ware of the interactive input: The function first performs PCA on each data block and then displays the eigenvalues (and a scree plot). Afterwards the function awaits the input from the user - it needs to know how many components need to be retained for that block.

References

Tenenhaus, A., & Tenenhaus, M. (2011). Regularized generalized canonical correlation analysis. Psychometrika, 76(2), 257-284.

Smilde, A.K., Mage, I., Naes, T., Hankemeier, T., Lips, M.A., Kiers, H.A., Acar, E., & Bro, R. (2016). Common and distinct components in data fusion. arXiv preprint arXiv:1607.02328.

Examples

 1 2 3 4 5 6 7 8 9 ## Not run: DATA1 <- matrix(rnorm(50), nrow=5) DATA2 <- matrix(rnorm(100), nrow=5) DATA <- cbind(DATA1, DATA2) R <- 5 Jk <- c(10, 20) pca_gca(DATA, Jk, cor_min = .8) ## End(Not run)

RegularizedSCA documentation built on May 2, 2019, 8:24 a.m.