Ridge Estimation of Precision Matrices from High-Dimensional Data

[Up] [Top]

Documentation for package ‘rags2ridges’ version 2.0

Help Pages

rags2ridges-package Ridge estimation for high-dimensional precision matrices
adjacentMat Transform real matrix into an adjacency matrix
conditionNumberPlot Visualize the spectral condition number against the regularization parameter
covML Maximum likelihood estimation of the covariance matrix
covMLknown Maximum likelihood estimation of the covariance matrix with assumptions on its structure
createS Simulate sample covariances or datasets
default.penalty Construct commonly used penalty matrices
default.target Generate a (data-driven) default target for usage in ridge-type shrinkage estimation
default.target.fused Generate data-driven targets for fused ridge estimation
edgeHeat Visualize (precision) matrix as a heatmap
evaluateS Evaluate numerical properties square matrix
evaluateSfit Visual inspection of the fit of a regularized precision matrix
fullMontyS Wrapper function
fused.test Test the necessity of fusion
GGMblockNullPenalty Generate the distribution of the penalty parameter under the null hypothesis of block-independence
GGMblockTest Test for block-indepedence
GGMmutualInfo Mutual information between two sets of variates within a multivariate normal distribution
GGMnetworkStats Gaussian graphical model network statistics
GGMnetworkStats.fused Gaussian graphical model network statistics
GGMpathStats Gaussian graphical model node pair path statistics
GGMpathStats.fused Fused gaussian graphical model node pair path statistics
hist.ptest Plot the results of a fusion test
is.Xlist Test if fused list-formats are correctly used
isSymmetricPD Test for symmetric positive (semi-)definiteness
isSymmetricPSD Test for symmetric positive (semi-)definiteness
kegg.target Construct target matrix from KEGG
KLdiv Kullback-Leibler divergence between two multivariate normal distributions
KLdiv.fused Fused Kullback-Leibler divergence for sets of distributions
loss Evaluate regularized precision under various loss functions
NLL Evaulate the (penalized) (fused) likelihood
NLL.fused Evaulate the (penalized) (fused) likelihood
optPenalty.aLOOCV Select optimal penalty parameter by approximate leave-one-out cross-validation
optPenalty.fused Identify optimal ridge and fused ridge penalties
optPenalty.fused.auto Identify optimal ridge and fused ridge penalties
optPenalty.fused.grid Identify optimal ridge and fused ridge penalties
optPenalty.LOOCV Select optimal penalty parameter by leave-one-out cross-validation
optPenalty.LOOCVauto Automatic search for optimal penalty parameter
pcor Compute partial correlation matrix or standardized precision matrix
plot.optPenaltyFusedGrid Print and plot functions for fused grid-based cross-validation
plot.ptest Plot the results of a fusion test
PNLL Evaulate the (penalized) (fused) likelihood
PNLL.fused Evaulate the (penalized) (fused) likelihood
pooledP Compute the pooled covariance or precision matrix estimate
pooledS Compute the pooled covariance or precision matrix estimate
print.optPenaltyFusedGrid Print and plot functions for fused grid-based cross-validation
print.ptest Print and summarize fusion test
rags2ridges Ridge estimation for high-dimensional precision matrices
ridgeP Ridge estimation for high-dimensional precision matrices
ridgeP.fused Fused ridge estimation
ridgePathS Visualize the regularization path
ridgeS Ridge estimation for high-dimensional precision matrices
rmvnormal Multivariate Gaussian simulation
sparsify Determine the support of a partial correlation/precision matrix
sparsify.fused Determine support of multiple partial correlation/precision matrices
summary.ptest Print and summarize fusion test
symm Symmetrize matrix
Ugraph Visualize undirected graph