Learn R Programming

singR (version 0.1.3)

Simultaneous Non-Gaussian Component Analysis

Description

Implementation of SING algorithm to extract joint and individual non-Gaussian components from two datasets. SING uses an objective function that maximizes the skewness and kurtosis of latent components with a penalty to enhance the similarity between subject scores. Unlike other existing methods, SING does not use PCA for dimension reduction, but rather uses non-Gaussianity, which can improve feature extraction. Benjamin B.Risk, Irina Gaynanova (2021) .

Copy Link

Version

Install

install.packages('singR')

Monthly Downloads

241

Version

0.1.3

License

MIT + file LICENSE

Maintainer

Liangkang Wang

Last Published

January 27th, 2025

Functions in singR (0.1.3)

orthogonalize

Orthogonalization of matrix
permTestJointRank

Permutation test with Greedymatch
gen.inits

Generate initialization from specific space
greedymatch

Greedy Match
permmatRank_joint

Permutation test to get joint components ranks
pmse

Permutation invariant mean squared error
lngca

Decompose the original data through LNGCA method.
%^%

Calculate the power of a square matrix
standard

Standardization with double centered and column scaling
singR

SImultaneous Non-Gaussian Component analysis for data integration.
whitener

Whitening Function
vec2net

Create network matrices from vectorized lower diagonals vec2net transfer the matrix vectorized lower diagonals into net to show the component image.
signchange

Sign change for S matrix to image
theta2W

Convert angle vector into orthodox matrix
tiltedgaussian

tiltedgaussian
create.graph.long

create graph dataset with netmat and mmp_order a data.frame called with vectorization of reordered netmat by mmp_order.
aveM

Average Mj for Mx and My Here subjects are by rows, columns correspond to components
NG_number

find the number of non-Gaussian components in the data.
covwhitener

Returns square root of the precision matrix for whitening
calculateJB

Calculates the sum of the JB scores across all components, useful for determining rho.
exampledata

Data for simulation example 1
angleMatchICA

Match the colums of Mx and My
est.M.ols

Estimate mixing matrix from estimates of components
curvilinear_c

Curvilinear algorithm based on C code with r0 joint components
curvilinear

Curvilinear algorithm with r0 joint components
matchICA

match ICA