This release addresses deprecation warnings of RcppArmadillo
pacotest 0.4.2
Updates and maintenance
The main reference was published in the Electronic Journal of
Statistics
Change maintainer mail address
pacotest 0.4.1
Updates, refactoring and bug
fixes
Major cleanup and refactoring of functionality without user
interface #38
pacotest is now also listed on r-universe
(https://maltekurz.r-universe.dev/ui#builds)
Fix bugs in the computation of the Ginv and Omega matrix #37 & #39
pacotest 0.4.0
Updates
Change of default parameters! By default the CCC test is now being
computed under consideration of estimation uncertainty of the
probability integral transforms, i.e., with options
withEstUncert = TRUE and
estUncertWithRanks = TRUE. Before, up to version 0.3.1,
both parameters defaulted to FALSE.
Note that when calling pacotest(U,W,'CCC'), the default
options for the CCC test are used (cf. pacotestset), but
the two parameters withEstUncert = FALSE and
estUncertWithRanks = FALSE are altered. In contrast when
calling pacotestOptions = pacotestset('CCC'), the two
parameters are set to withEstUncert = TRUE and
estUncertWithRanks = TRUE. For the CCC test, under the
default setting, it is assumed that estimated PPITs are provided and the
test statistic is computed under consideration of estimation uncertainty
of the probability integral transforms, i.e.,
withEstUncert = TRUE and
estUncertWithRanks = TRUE. To apply pacotest
with withEstUncert = TRUE, three additional inputs have to
be provided (data, svcmDataFrame and
cPitData).
In the vine copula context, PPITs are usually estimated and not
known. Therefore, in the vine copula context it is recommended to use
the functions pacotestRvineSeq or
pacotestRvineSingleCopula instead of pacotest.
These functions automatically pass through the additional arguments
data, svcmDataFrame and cPitData
to the function pacotest and the CCC test can be applied in
its default setting with consideration of estimation uncertainty of the
probability integral transforms, i.e., withEstUncert = TRUE
and estUncertWithRanks = TRUE.
Continuous integration is now done with github actions
(https://github.com/MalteKurz/pacotest/actions) instead of travis and
appveyor.
Minor improvements and bug
fixes
Fixed a couple of typos in the documentation.
Updated the reference to Spanhel, F. and M. S. Kurz (2019),
“Simplified vine copula models: Approximations based on the simplifying
assumption”, Electronic Journal of Statistics 13 (1), 1254-1291.
pacotest 0.3.1
Updates
The default method for generating from a discrete uniform
distribution changed (R version >=3.6.0). Regression test results
have been adapted accordingly.
pacotest 0.3
Updates
Renaming of ECORR test to CCC test to be
in line with the corresponding paper (Kurz and Spanhel (2017) https://arxiv.org/abs/1706.02338)
Added an additional, more informative, output,
testResultSummary, to pacotestRvineSeq()
Option, stopIfRejected, added to
pacotestRvineSeq(), which allows the user to stop the
sequential test procedure in case of a rejection
Usage of Bonferroni correction in
pacotestRvineSeq()
Default value of aggInfo is now set to
meanAll to be in line with the paper
Minor improvements and bug
fixes
Bug fix in extractSubTree; Added a corresponding unit
test
Stabilization of numerical derivatives in edge cases for the copula
parameters
Don’t use the aggregated information for computing the test
statistic with the Gamma0 partition
Fixed an edge case becoming relevant when almost all copulas in a
vine copula are set to independence copulas
pacotest 0.2.2
Bug fixes
#24
removed calls of floating-point function floor on integers (caused
installation failures on solaris)
#23
prevent nan’s in unsigned int, which is outside the range of
representable values (caused memtest note on CRAN)
pacotest 0.2.1
Bug fixes
#14
removed calls of floating-point functions on integers (caused
installation failures on solaris)
#15
prevent nan’s in unsigned int, which is outside the range of
representable values (caused memtest note on CRAN)