DESCRIPTION: ISBN changed to <urn:isbn:978-3838108148> (see mail from Kurt Hornik from 06.06.2023)
tests/thubert.Rout.save, tlda.Rout.save, tmcd4.Rout.save, tqda.Rout.save: Follow up to changes in robustbase 0.99-0
data/ionosphere.rda - data set added
R/CovRobust, plot_utils: 'xyqqchi2' and 'xydistance' plots, fixed to use of parameter id.n
Fixed - warning: invalid UTF-8 in comment [-Winvalid-utf8] in src/rrcov_init.c
NOTHING NEW YET
Fixed - maxcond passed to the low level functions
Fixed errors "Found if() conditions comparing class() to string" in detmrcd.R and plot_utils.R
Fixed NOTE "Found the following math rendering problems" in PcaHubert.Rd, see mail from Kurt Hornik from 11 August 2022
Fix for USE_FC_LEN_T becoming the default in 4.2.0
Fortran warnings fixed in ds11.f and fsada.f
Examples for PCA/adjustment for skewed data and PCA/percentage of explained variance added
minor differences in tests for PcaProj() on some platforms fixed
scoreplot() corrected to show the labels of the samples
Data set Fruit added: fruit.rda
URLs in Rd files replaced by DOIs to fix for the migration of the www.jstatsoft.org to a new editorial system (see mail from Achim Zeileis from 06.10.2021)
Fixed a problem when showing the percentage of explained variance in summary() of all PCA functions when k is chosen to be less than the number of variables in the input data matrix (k < p). The returned object contains now additionally the rank of the input matrix, the original eigenvalues (of all variables) and the original total variance, if available.
PcaHubert: option for adjusted outlyingness for skewed data added
Data set Computer Hardware added: machines.rda
Data set Wolves added: wolves.rda
Fixed some URLs, particularly the reference to javasoft.org
PcaHubert will crash if X is 1-dimensional and mcd=FALSE: fixed.
Fixed 'noLD' issues in tlda.R and okg4.R
Fixed a problem with wrong scores in PcaProj() (reported by Matthieu Lesnoff <firstname.lastname@example.org>)
Fixed a problem with nsamp="exact" or nsamp="best" in CovMve(), CovSest() (reported by Claudio Agostinelli) - these functions, differently from CovMcd, should not take non-numeric 'nsamp'
Added parameter 'control' to Linda - to select the robust location and covariance estimator to use in LDA. Now any estimator derived from class 'Cov' can be used, even such that are not in 'rrcov'. Return this parameter in the returned S4 object.
Linda returns now the location and covariance estimator used as 'covobj'. This is useful for controlling cross-validation, for example.
Linda and LdaClassic use generalized inverse if the common covariance matrix is singular.
Fixed an issue the 'predict' function.
Removed the dependence on packages 'cluster' and 'ellipse'.
Added data set
diabetes; data set
soil from package
rrcovHD moved here.
Linear and quadratic discriminant analysis can use the MRCD estimates.
Fixed an issue with CovControlMcd(nsamp="deterministic") - this would not work, because nsamp was defined in the class definition as "numeric". Now it is "Cnumeric" - union of "character" and "numeric'.
Corrected the code for Minimum Regularized Covariance Determinant estimator (MRCD) -
CovMrcd() - the step of adjusting the eignevalues in r6pack() is excluded
now because it has no effect when n > p.
Added Minimum Regularized Covariance Determinant estimator (MRCD)
(Boudt et al. 2018) -
Added data set
octane; data set
olitos from package rrcovHD moved here.
The 'pairs' plot is now available for classical covariance matrix