Release Date: 2013-09-18
- >=2.14 for version 2.5.0
- >=2.6.0 for version 2.1,
- >=2.2.0 for versions 0.4-2 -0.4-4 and 1.8-2.0,
Dependencies: requires packages "methods"
available from CRAN
as well as package "distr"
version >= 1.9 by the authors of this package and "startupmsg
" by Peter
also available from CRAN
What is "distrEx" meant for?
This package provides some extensions to package "distr"
- extreme value distribution classes (to be moved to new package "RobExtremes" from version 2.4 on),
- in the form E(X) for the expectation of X where X is some
- in the form E(X,f) for the expectation of f(X) where X is
some distribution and f some function in X,
- further functionals: var, sd, median, IQR, mad
- truncated moments,
- distances between distributions (Hellinger, Kolmogorov, total
variation, "convex contamination")
- lists of distributions
- conditional distributions in factorized form
- conditional expectations in factorized form
more on that package may be found in Matthias' PhD Thesis "Numerical Contributions to the Asymptotic
Theory of Robustness"
, App. D
available on http://stamats.de/ThesisMKohl.pdf
as well as in an extensive manual
distrXXX packages (up to version 0.4-3)
0.4-3 on, we have converted this manual into a common vignette to
, distrMod, distrTeach
which is available in the mere documentation package distrDoc
. To use it you may type
require("distrDoc"); V<-vignette("distr"); print(V); edit(V)
This software is distributed under the terms of the GNU GENERAL
PUBLIC LICENSE LGPL v3,
- included into the .tar.gz.file
also see demo(package="distr") and demo(package="distrEx") --- after installation of "distr" and "distrEx"
an illustration of the use of these functionals in programming
a deconvolution expample: let X and E be independet univariate random
variables with [almost] arbitrary, given absolsutely continuous
We generate the (factorized) conditional distribution of X given
Changes from Version 0.4-1 to Version 0.4-2
- ContaminationSize, HellingerDist, KolmogorovDist, TotalVarDist
now return a list which consists of the corresponding distributions and
- minor changes in m1df and m2df to increase speed of computation
- minor changes in DiscreteMVDistribution to increase speed of
- introduction of a new parameter useApply in methods for function
with default value TRUE
Changes from Version 0.4-2 to Version 0.4-3
- Implementation of functionals:
- Internationalization: use of gettext, gettextf in output
- C-interface .GLaw() to replace respective R-Code in
- Inclusion of demos (see above)
Changes from Version 0.4-3 to Version 0.4-4
Changes from Version 0.4-4 to Version 1.8
- exact versions for var() + E() for DExp
- dim method for DiscreteMVdistributions
Changes from Version 1.8 to Version 1.9
- besides minor changes /typo fixing essentially: changed
version counter to the one for the other distrXXX-packages
Changes from Version 1.9 to Version 2.0
- made calls to uniroot(),
integrate(), optim(ize)() compliant to https://stat.ethz.ch/pipermail/r-devel/2007-May/045791.htmll
- 'distrEx' now
behaves exactly the same as the other members of the distrXXX family as to 'distrExOptions()', 'getdistrExOption()'
- substantial contributions by Jay
& 'kurtosis()' are
now available as functionals
- 'E()', 'var()'
return 'NA' in case of
T-distribution if not defined
- some exact formulas for 'median()',
'mad()', and 'IQR()'
- moved some parts from
from package 'distrEx'
to package 'distr'
- generating function 'DiscreteDistribution'
- univariate methods of 'liesInSupport()'
- classes 'DistrList' and 'UnivariateDistrList'
- generating functions EuclideanSpace() ,Reals(), Naturals()
- new function 'illustrateLLN' to
illustrate LLN and demo 'illustLLN'
- similar preset strings substituted in both expression and
vectors as in plot-methods
from package distr
- enhanced 'illustrateCLT'
- 'illustrateCLT' no longer is
a generic function but a regular function
- the plotting feature of 'illustrateCLT' is extracted
and has become
a generic function 'plotCLT'
(now with title and the summands mentioned
in the header)
- there is a TclTk-based demo now (therefore
TclTk is a suggested package now)
- plot includes a title
- for 'DiscreteDistributions'
in the "d"-panel, the support is thinned out if length too long
- Komogoroff-distance is printed out
- replaced recursive summation in illustrate-CLT method by 'convpow()'
(for summand distribution of class 'LatticeDistribution' / 'AbscontDistribution')
- new file MASKING and corresponding command 'distrExMASK()' to
describe intended/non-intended maskings
- mentioned in package-help: startup messages
may now also be suppressed by
(from package 'base')
- formals for slots p,q,d
as in package stats to
- p(X)(q, [cond,]
TRUE, log.p = FALSE)
- q(X)(p, [cond,]
TRUE, log.p = FALSE)
- d(X)(x, [cond,] log =
- used wherever possible;
but backwards compatibility: always checked whether lowert.tail / log / log.p are
- E-, var-, IQR-, mad-, median-, kurtosis-, skewness- methods
for new class union 'AffLinDistribution'
- moved all teaching illustrations for LLN and CLT to new package "distrTeach"
- E methods for class[es]
- distance methods for class
- CvM distance is implemented
Changes from 2.0 to
- gains distribution Pareto; ported from pkg actuar by Nataliya
- with corresponding functionals E, var, IQR, median, skewness,
- new slots d,p,q for Gumbel distribution catching errors,
vectorization and lower.tail, log[.p] argument ...
- E, var, IQR, median, skewness, kurtosis for Gumbel
- new methods for Gumbel distribution ..
- expectation gains (optional) arguments low and upp for lower
and upper bounds of the integration range;
- other functionals based on this may use these arguments,
too (var, sd, skewness, kurtosis)
- expectation gains explicit arguments to set accuracy locally;
- also both quantile and scale based methods is used to
determine a sensible integration range in expectation
- median and IQR are now defined for UnivariateCondDistribution
- general bug fixes:
- checked and fixed functionals (stirred up by mail by Jay
- bug corrected in E for Hyper (thanks to Jay G. Kerns!) ...
- corrected small bug in mad, added new implementation for
skewness and kurtosis for signature "ANY".
- corrected small bug in skewness for AffLinDistribution.
- introduced distance OAsymTotalVarDist (minimal asymmetricctotal
- introduced new asymmetric total variation distance
- TotalVarDist and HellingerDist gain extra arguments to better
control the integration range and exactness
- NEW FUNCTIONALS
- m1df for AffLinDistribution
- Expectation, var, IQR, median, skewness, kurtosis available for
... and: had forgotten to document the plot-methods in distrExsome
typo's in integration range selection
- several buglets detected with the fuzzier checking mechanism
- cf [Rd] More intensive checking of R help files,
Prof Brian Ripley, 09.01.2009 10:25)
- [Rd] Warning: missing text for item ... in \describe? , Prof
- LICENSE: moved license to LGPL-3
Changes from 2.1 onwards
see NEWS file
Our plans for
the next versions:
This page is
maintained by Peter
and last updated on 2013-09-18.