Version 1.4-0

New features

The Peto-Peto test for right-censored data is now available (see ?logrank_test and ?logrank_weight).
The Moore-Penrose inverse of the conditional covariance matrix can now be extracted (see ?covariance).
For tests of conditional independence within blocks, i.e., stratified tests, the partial linear statistic for each block and its conditional expectation and variance as well as its conditional covariance and corresponding Moore-Penrose inverse can now be extracted (see ?statistic, ?expectation, ?variance and ?covariance).

Exact multivariate two-sample tests with quadratic test statistics appeared to work but returned incorrect p-values; it now results in an error.
Extraction of the standardized linear statistic from objects of class "IndependenceLinearStatistic" resulted in an error.

Extraction of the conditional expectation or variance now returns a matrix instead of a vector (for consistency with linear statistics and p-values).
The test size is now computed more efficiently.
Documentation updates.

The internal storage of the linear statistic and its conditional expectation and covariance, i.e., slots linearstatistic, expectation and covariance in classes inheriting from "IndependenceLinearStatistic", as well as the Moore-Penrose inverse of the conditional covariance, i.e., slot covarianceplus in classes inheriting from "QuadTypeIndependenceTestStatistic", has been changed. In particular, the conditional covariance and its Moore-Penrose inverse now use a memory-efficient packed format.
R >=3.6.0 and package libcoin >=1.0-5 are now required.