CovCorTest1.1.0 package

Statistical Tests for Covariance and Correlation Matrices and their Structures

ATS_fun

ATS for transformed vectors

ATS

Anova-Type-statistic

ATSwS

Anova-Type-Statistic with weighted sum

Bootstrap_trans

Bootstrap using transformation for one group

Bootstrap

Bootstrap for one and multiple groups

CombTest

CombTest Object

CovTest

CovTest Object

dvech

Diagonal vectorization

generateData

Function to generate bootstrap observations

get_extended_matrix

Extend a matrix to full rank using identity columns

get_hypothesis

Construct hypothesis matrix and vector from linear covariance model st...

Jacobian

Jacobian matrix for transformation functions

Listcheck

Function to transform the data into a list, if there are not already

print.CombTest

Print function for CombTest object

print.CovTest

Print function for CovTest object

Qvech

Auxiliary function to calculate the covariance of the vectorized corre...

subdiagonal_mean_ratio_cor

Root transformation of the vectorized correlation matrix

subdiagonal_mean_ratio_fct

Transformation of the vectorized covariance matrix by quotients of mea...

Tayapp1G

Function for the Taylor-based Monte-Carlo-approximation for one group

TayappMG

Function for the Taylor-based Monte-Carlo-approximation for multiple g...

TaylorCombined

The Taylor-based Monte-Carlo-approximation for a combined test

test_combined

Combined test for equality of covariance matrices and correlation matr...

test_correlation_structure

Test for structure of data's correlation matrix

test_correlation

Test for Correlation Matrices

test_covariance_structure

Test for structure of data's covariance matrix

test_covariance

Test for Covariance Matrices

vdtcrossprod

Function to calculate dvech(X t(X))

vechp

Vectorization of the upper triangular part of the matrix

vtcrossprod

Function to calculate vech(X t(X))

WDirect.sumL

Weighted direct sums for lists

A compilation of tests for hypotheses regarding covariance and correlation matrices for one or more groups. The hypothesis can be specified through a corresponding hypothesis matrix and a vector or by choosing one of the basic hypotheses, while for the structure test, only the latter works. Thereby Monte-Carlo and Bootstrap-techniques are used, and the respective method must be chosen, and the functions provide p-values and mostly also estimators of calculated covariance matrices of test statistics. For more details on the methodology, see Sattler et al. (2022) <doi:10.1016/j.jspi.2021.12.001>, Sattler and Pauly (2024) <doi:10.1007/s11749-023-00906-6>, and Sattler and Dobler (2025) <doi:10.48550/arXiv.2310.11799>.

  • Maintainer: Svenja Jedhoff
  • License: GPL (>= 3)
  • Last published: 2025-10-17