                        Changes in version 1.1-8                        

Misc

    o   No longer use the newly deprecated structure(.Names = *).

    o   Fix some URLs.

                 Changes in version 1.1-7 (2026-02-20)                  

New Features

    o   New experimental gofMMDtest(), multiplier bootstrap, based on
	the squared MMD.

    o   Add volume() as an auxiliary for prob(), as an S3 method for
	the volume() generic from cluster.

Bug Fixes

    o   mvdc() more carefully checks paramMargins, and now warns when
	it contains named vectors instead of lists, thanks to Rolf
	Turner's e-mail (in 2015).

    o   loglikMvdc() now calls dMvdc(.., log=TRUE) instead of
	log(dMvdc(..)).  It and fitMvdc() get new options for (not)
	signalling warnings.

    o   fitMvdc() consequently works on the log scale instead of taking
	logs; this is typically faster and more accurate, but notably
	even necessary for “large” n or large dim= d.

    o   Fix the _relative_ link to the ‘Frank-Rmpfr’ vignette in
	help(copula-package).

    o   Update URLs in the vignettes.

                 Changes in version 1.1-6 (2025-03-18)                  

Bug Fixes

    o   Validity checking of "acopula"-classed objects no longer
	requires psi(0, theta = 1/2) = 1, but only approximate
	equality.

                 Changes in version 1.1-5 (2025-02-09)                  

New Features

    o   The density, and hence likelihood and fitCopula() fits for
	Clayton copulas has been robustified numerically, such that the
	log likelihood does not underflow to -Inf unnecessarily.

Bug Fixes

    o   fitCopula(.., method = "Nelder-Mead") no longer shows optim()
	warnings (in 1d-case).

Misc

    o   system.time() results in ../tests/Stirling-etc.R are no longer
	tested (unless doExtras), as they can't work e.g., for valgrind
	extra tests.

    o   Speed up 3 more examples.

                 Changes in version 1.1-4 (2024-08-17)                  

Bug Fixes

    o   fgmCopula(param, dim) now checks length(param) correctly.

Misc

    o   Mention how to use colors in ?wireframe2.

    o   Regr.tests e.g. with valgrind are really slow; hence, do no
	longer check for system.time() used (unless doExtras)

                 Changes in version 1.1-3 (2023-12-07)                  

Misc

    o   format fix ("%lld"); use C99 SIZE_MAX

    o   fix "lost braces"

    o   nacopula-pkg.Rnw fix for latex

    o   Avoid false positive warning: generic sign() vs "method"
	sign.2rotC().

                 Changes in version 1.1-2 (2023-01-25)                  

Miscellaneous

    o   Avoid long lines in vignette R output.

    o   Replace sprintf() by snprintf() as it is going to be deprecated
	in upcoming C environments, e.g., on macOS 13.

    o   Fix _all_ current compiler warnings

                 Changes in version 1.1-1 (2022-11-17)                  

Miscellaneous

    o   Fix non-UTF-8 accents in C code (‘WARN’ on CRAN).

    o   Decrease the large inst/doc/*.html vignettes a bit more, for a
	smaller package size (for CRAN).

                 Changes in version 1.1-0 (2022-06-15)                  

New Features

    o   fitCopula(<mixCopula>) now by default uses estimate.variance =
	TRUE also with _free_ mixture weights.

    o   fitLambda() gets new method "Schmidt.Stadtmueller".

    o   fitMvdc() works in more cases, e.g. with a mixCopula.

    o   "mvdc" class objects no longer inherit from "xcopula" (a
	copula) but from new "Xcopula" (which just _contains_ a
	copula).

    o   using isTRUE() and isFALSE() instead of identical(TRUE, .) etc.

    o   fitCopula(.., method="mpl") now also catches errors in var-cov
	matrix computations.

    o   New loglikCopulaMany(pList, ..) function for (more efficiently)
	computing copula likelihoods for many parameter vectors /
	values.

    o   For elliptic (t-, normal-) copulas, provide “permuted Toeplitz”
	dispersion via new dispstrToep() which allows to specify a
	_permutation_ of the d variables / columns / coordinates to
	have a Toeplitz correlation structure.

Bug Fixes

    o   New arguments and defaults keepAttr=FALSE and checkCorr=FALSE
	for mvtnorm's Miwa() should speed up pCopula() for normal- and
	t- copulas.

    o   Empirical copula computations tweaks, preventing some erronous
	computations.

                 Changes in version 1.0-1 (2020-12-12)                  

New Features

    o   The pCopula() methods for normal- and t-copulas now allow to
	pass algorithm = . to pkg mvtnorm's functions.  For lower
	dimensions, notably dimension d <= 3, the default is now to
	typically use a _non_-random algorithm.
	
	Such a default algorithm is also used in the (internal)
	derivative for normal- and t-copulas, (dCdu() calling
	dCduEllipCopula()).

    o   invisible improvement in internal Jscore() for internal
	var.mpl() computations.

    o   For mixCopulas, the default optimMethod() for fitCopula() is
	now "L-BFGS-B".

    o   When fitCopula(*, estimate.variance=TRUE) fails during optim(*,
	hessian=TRUE), a warning() is signalled, and kept as attribute
	of the NA-valued vcov matrix.
	
	Similarly when the hessian inversion fails, the warning message
	is now returned with the NA-valued vcov matrix.

    o   Smoothed empCopula() objects now have the right sampling
	methods.

Bug Fixes

    o   fitCopula(*, estimate.variance=FALSE) now returns a 0 x 0
	matrix as slot @ var.est in all cases.

    o   tweaks some of the regression tests.

    o   For mixture copulas (mixCopula(.)) estimation by fm <-
	fitCopula(*, estimate.variance=TRUE) now is more reasonable:
	summary(fm) and vcov(fm) still stay in the w-space and
	currently return a warning about not-yet implemented Jacobian
	delta-method.  coef(), vcov(), and summary(fm) now all have an
	optional argument orig = TRUE, which when set to false will
	provide results on the ‘l-scale’ aka ‘lambda-scale’ (instead of
	the original ‘w-scale’) for the weights, such that inference on
	the ‘l-scale’ is available.

                 Changes in version 1.0-0 (2020-05-29)                  

New Features

    o   mixCopula() objects can now be fit (by fitCopula()) more
	generally, as the weights no longer need to be kept fixed,
	thanks to Maechler's clr1() parametrization, and Rebecca
	Morger's B.Sc thesis.

    o   When no start parameter vector is specified in fitCopula(), the
	new (generic) function getIniParam() is used now; it notably
	has a method for "mixCopula" copulas.

    o   In the fitCopula*() functions, the traceOpt option now can be a
	positive integer (instead of just TRUE and FALSE) to print
	trace info only every traceOpt objective function evaluation.

Bug Fixes

    o   corrected thinko in `freeParam<-`(<mixcopula>, v).

    o   more tweaks on fixed / free parameter settings for "mixCopula".

                Changes in version 0.999-20 (2020-02-06)                

New Features

    o   evTestA() got new options trace.lev and report.err for
	reporting numerical intergration problems.

    o   gofT2stat() implements the two-sample test statistic of
	Rémillard and Scaillet (2009).

Bug Fixes

    o   rCopula() of an empCopula object failed because sampling was
	done without replace = TRUE.

    o   cCopula() now should work for indepCopula() and "moCopula"
	objects (via internal rosenblatt().

    o   fix src/empcop.h and man/fgmCopula.Rd inconsistencies, found by
	latest checks on CRAN.

                Changes in version 0.999-19 (2018-12-21)                

New Features

    o   exchTest() gets new argument ties.method.

    o   a new vignette HAXC on Hierarchical Archimax Copulas.

    o   New simple classes and constructor fhCopula() for lower and
	upper Frechet-Hoeffding bounds (W and M), see ?fhCopula.

    o   Empirical copulas now get an explicit class and constructor
	empCopula().

    o   Marshall-Olkin copulas for the bivariate case (d = 2) get a
	class representation and are constructed via moCopula().

    o   In the case where loglikCopula(start, u=u, copula=copula) is
	+Inf, as e.g., in the case of data with duplicate columns
	(perfect correlation), the internal fitCopStart() and hence
	fitCopula() tries harder to work with the iTau()-generated
	start values.

    o   The plot() method for copula objects uses a somewhat
	experimental smart default for the main title.

    o   The "copula" class now gets its dimension slot from the new
	virtual class "dimCopula" which contains all copulas with an
	explicit dimension slot.

    o   the "indepCopula" now extends only "dimCopula" and "parCopula"
	and hence no longer has unuseful slots such as parameter.

    o   contributors now are all mentioned in DESCRIPTION; and all are
	listed including their ORCID's.

    o   gofCopula() gets new argument test.method.

    o   cCopula() now works for rotCopula objects and bivariate
	mixCopula objects.

Bug Fixes

    o   Bugfix in dCopula() method of empCopula(*, smoothing = "beta").

    o   Better error messages when copula parameters are NA (and hence,
	e.g. pCopula() gives an error).

    o   Bugfix in rCopula(, copula = mixCopula()) if the weights are
	unit vectors.

    o   mvdc(<rotCopula>, *) now works, too.

    o   fixed thinko in internal chkFun() called e.g., by perspMvdc().

                Changes in version 0.999-18 (2017-09-01)                

New Features

    o   contourplot2() now gets argument pretty, to be passed on, with
	a better (but not back-compatible) default (!labels).

    o   new margCopula() constructor of marginal copulas, currently for
	normal-, t-, and Archimedean copulas only.

Bug Fixes

    o   cCopula(*, inverse=TRUE) now also works for non-Clayton
	Archmedean copulas.

                Changes in version 0.999-17 (2017-06-18)                

New Features

    o   C.n() and several tests now get an optional argument
	(ties.method or similar) to specify what happens with ties.

    o   setTheta(cop, val) now also works for a partially fixed
	parameter elliptical copula.

    o   Export the (formerly hidden) utilities isFree(), isFreeP(), and
	nParam().

    o   New rLatinHypercube() and rAntitheticVariates() utilities.

    o   dCopula(u, copula, *) now signals an error if ncol(u) and
	dim(copula) differ.

    o   corKendall() now checks for NA's by default and then works
	pairwisely.  It is now documented and exported.  As a
	consequence, etau(U, *) now also works in the case U has
	missing values.

Bug Fixes

    o   In the case of _ties_, several GOF tests now use a better
	default method for dealing with ties in their calls to pobs().

    o   pobs(<zoo>, .) continues working.

    o   setTheta(<tCopula>, ..) now works correctly with free/non-free
	parts of the parameters.

    o   fgmCopula() fixed to work with rotCopula().

    o   iTau() fixed for joeCopula().

    o   Improved warning messages in iTau() and iRho().

    o   rCopula(<frankCopula>) now also works for the case of large tau
	and d > 2 (via re-parametrized log series distribution).

                Changes in version 0.999-16 (2017-01-09)                

Bug Fixes

    o   pK(u, log.p=TRUE) now correctly checks u.

    o   rotCopula(cop) now _does_ rotate in the default case.

    o   vcov(fitCopula(*, "mpl")) now works (returning a matrix of
	NAs).

    o   fitCopula(joeCopula(), *, method = "itau") gave an error
	instead of warning.

    o   setTheta(<rotCopula>, <par>) now works.

    o   xvCopula(<rotCopula>, <data>) now works.

    o   fitMvdc() now works again for a “df.fixed-tCopula”.

    o   "mvdc" methods for contour(), persp() and wireframe2() now use
	ylim correctly.

    o   cCopula(<archmCopula>) failed for 2-dimensional AMH, Clayton,
	and Frank in the _negative_-tau case.
	
	Also, cCopula() sometimes returned a vector instead of an n * k
	matrix (which ?cCopula has promised always).

    o   pCopula() was wrong (but gave a warning) for some (unbalanced)
	nested archimedean copulas.

New Features

    o   Adding an abstraction level, using (currently hidden) "setr"
	and "getr" methods, allows to simplify class definition for
	"khoudrajiCopula" and fitCopula() now to also work e.g. for
	rotCopula(khoudrajiCopula(.), ..).

    o   The coef() method for fitCopula() or fitMvdc() results can get
	the matrix of parameters _and_ their standard errors by the new
	option SE = TRUE.

    o   summary(fitCopula(*)) and summary(fitMvdc(*)) have a nicer
	print() method and print _more_ than just printing the fitted
	objects.  Note that

    o   printing fitCopula(*) or fitMvdc(*) no longer computes and
	prints standard errors ("Std. Error"), but printing
	summary(fit..(*)) does (see above).

    o   improved documentation about pK() and qK().

    o   mixCopula() and "mixCopula" class for arbitrary finite mixtures
	of copulas.

    o   splom2() now works when a single “global” color is specified.

                Changes in version 0.999-15 (2016-07-25)                

    o   C.n(u, X, *) now has X as second argument, calls pobs(X) and
	then F.n(*,*); previously, C.n and F.n were practically
	equivalent, which has been confusing.

    o   pobs() is more flexible, accepting numeric vectors, and e.g.,
	keeping "xts" and "zoo" class attributes.

    o   improved R markdown vignettes.

    o   Internally, notably for etau() and "itau" fitting, make use of
	fast Kendall's tau from package pcaPP (via hidden
	corKendall()).

    o   fitCopula() gets now method "itau.mpl" for tCopulas, which
	implements the robust estimation method of Mashal and Zeevi
	(2002).

    o   various improvements in gofCopula() related to the alternative
	test statistics (method != "Sn").

    o   new xvCopula() function: cross-validated copula fitting for
	model selection.

    o   Asymmetric copulas are now better implemented; previous
	asymCopula() and asymExplicitCopula() are deprecated and to be
	replaced by khoudrajiCopula().

    o   “Rotated” copulas, i.e., generalizations of “survival” copulas,
	are now available via rotCopula().

    o   gofEVCopula()'s default optim.method has been changed from
	"Nelder-Mead" to "BFGS" in line with fitCopula() and because
	Nelder-Mead is warned against in the 1-dimensional case.

    o   new getAname() useful e.g., in htrafo().

    o   Parts of a parameter vector can be kept fixed (“known”), see
	fixedPar.  The "tCopula" and "tevCopula" classes consequently
	no longer have a df slot.

    o   Functions deprecated (.Deprecated) since 2012, namely, K,
	beta.hat, calibKendallsTau, calibSpearmansRho, kendallsTau,
	spearmansRho, genInv, genFun, genFunDer1, genFunDer2, Afun,
	AfunDer, Anfun, dcopula, pcopula, rcopula, dmvdc, pmvdc, and
	rmvdc are now defunct (the user still sees them and gets a
	useful error message from .Defunct()!).

    o   Function tailindex is deprecated in favor of new lambda();
	gofMB() and gofPB() calls are deprecated and to be replaced by
	gofCopula(.., simulation = *).

    o   Function rtrafo() has now been deprecated, users should work
	with cCopula() instead.

    o   Various improvements in data visualization tools such as
	contour(), persp(), contourplot2(), pairs2(), wireframe2(),
	cloud2() and splom2().

    o   qK() gets log.p option.

                Changes in version 0.999-14 (2015-10-26)                

    o   New demo "NALC" on nested Archimedean Levy copulas

    o   Explicitly import from "base packages"

    o   contour(<mvdc>, ..) gets new default box01=FALSE, as a [0,1]^2
	box does not make sense for general multivariate distributions.

    o   coeffG() is exported (though still partly subject to change)

    o   new (rmarkdown -> knitr) vignette wild_animals from former
	demo.

    o   pCopula(*, cop) for normal- or t- copulas cop now accepts
	optional arguments to be passed to the underlying (numerical
	integration) algorithm from package mvtnorm.

    o   fixed the (dontrun) 3d example of gofCopula.

                Changes in version 0.999-13 (2015-03-05)                

    o   New demo on tail compatibility

    o   lsum(): also work for vectors and when result is zero, i.e.
	log(0) == -Inf.

                Changes in version 0.999-12 (2014-10-02)                

    o   Fix bug which did not allow to set non-default optim.method in
	gofCopula().

    o   Not fully back-compatible change: gofCopula() now passes all
	extra arguments via ... to fitCopula().  Before, gofPB() passed
	them to the trafo.method.

                Changes in version 0.999-11 (2014-09-05)                

    o   The default for the optional estimate.variance argument of
	fitCopula() and fitMvcd() now is TRUE only if the optimization
	converged.

    o   rtrafo() is more efficient for "normalCopula".

    o   fitMvdc() now also works when margin parameter has length >1.

    o   fitMvdc() gains optional arguments lower and upper for
	specifying box-constraints to optim().

                Changes in version 0.999-10 (2014-06-19)                

    o   Fixes some border cases, notably psi() for negative tau Gumbel,
	Clayton and AMH copulas,

    o   debye[12]() functions improvements (e.g., Inf)

    o   New log1pexp() from Rmpfr's vignette log1mexp.

    o   rho(<amhCopula>) now numerically stable, also works for theta
	~= 0, using Martin's beautiful formula, as documented in the
	_new vignette_ rhoAMH-dilog.Rnw.

    o   Purely numerical .psiFrank() etc; accuracy improvements for
	psi(), rho(), tau() for small |alpha| for Frank.

    o   New cospi2(), tanpi(); iTau() and iRho() have a smaller
	tolerance of 1e-7 and hence are more accurate where they use
	inversion.

    o   polylog(x, 2) now uses gsl's dilog().

    o   pCopula() now works for a t-copula with df=Inf.

                Changes in version 0.999-9 (2014-05-05)                 

    o   rtrafo() with new option 'inverse=TRUE', hence can be used for
	sampling.

    o   Fixed overflow bug (seg.fault) for large n^2 p

    o   Fixed qqplot2() case

                Changes in version 0.999-8 (2014-02-04)                 

    o   pobs(*, lower.tail) new optional argument.

    o   new optional argument indepC.maybe to (Archimedean) copula
	constructors.

    o   empirical copula functions C.n() and F.n()

    o   all.equal(*, tol) adaption to upcoming R changes.

                Changes in version 0.999-7 (2013-05-21)                 

    o   a bug fix version, as 0.999-6 had in one case accidentally
	introduced a *NON*-back compatible version in gofCopula().

                Changes in version 0.999-6 (2013-05-14)                 

    o   gofCopula() gets a new trafo.method argument, now also can use
	(new) htrafo() in addition to Rosenblatt's transform rtrafo.

    o   ...

                Changes in version 0.999-5 (2012-12-03)                 

    o   new back-compatibility feature which allows older R scripts
	(notably the 2010 JSS paper) to still run without errors.

                Changes in version 0.999-4 (2012-11-16)                 

    o   ...

                Changes in version 0.999-3 (2012-10-27)                 

    o   ...

                Changes in version 0.999-2 (2012-10-25)                 

    o   ...

                Changes in version 0.999-1 (2012-08-13)                 

    o   ...

                Changes in version 0.999-0 (2012-07-30)                 

    o   Renaming (and deprecation) “orgy”: ... ...

    o   ...

    o   ...

    o   ...

                 Changes in version 0.99-4 (2012-07-03)                 

    o   dK(), pK(), qK(), rK(): Kendall's distribution functions.

    o   ...

                 Changes in version 0.99-2 (2012-05-30)                 

    o   export log1mexp()

    o   added facilities for Hofert & Maechler's graphical GOF testing.

    o   fixed corner cases of dcopulat() and dcopulaNormal().

    o   polyG() and dsumSibuya() improvements; possibly using Rmpfr.

    o   fixed "boundary overrun" bug in indepence tests built on
	empirical copula.

                 Changes in version 0.99-0 (2012-03-30)                 

    o   Big merge of former CRAN packages nacopula (Maechler and
	Hofert) and copula (Yan and Kojadinovic) into new “super
	package”.

    o   new class "Copula", with [dpr]copula() methods for "nacopula".
