[R-pkgs] New CRAN package `coin'
Torsten.Hothorn at rzmail.uni-erlangen.de
Fri Jun 3 08:59:12 CEST 2005
Conditional Inference Procedures in a Permutation Test Framework
The `coin' package implements a general framework for conditional
inference procedures, commonly known as permutation tests,
theoretically derived by Strasser & Weber (1999). The conditional
expectation and covariance for a broad class of multivariate linear
statistics as well as the corresponding multivariate limiting distribution
was derived by Strasser & Weber (1999). These results are
utilized to construct tests for independence between two sets of
Beside a general implementation of the abstract framework the package
offers a rather huge set of convenience functions implementing well
known classical as well as less prominent classical and non-classical test
procedures in a conditional inference framework. Examples are linear rank
statistics for the two- and K-sample location and scale problem against
ordered and unordered alternatives including post-hoc tests for arbitrary
contrasts, tests of independence for contingency tables, two- and K-sample
tests for censored data, tests for independence of two continuous
variables as well as tests for marginal homogeneity and symmetry.
Conditional counterparts of most of the classical procedures given in
famous text books like Hollander & Wolfe (1999) or Agresti (2002) can be
implemented as part of the general framework without much effort.
Approximations of the exact null distribution via the limiting
distribution and conditional Monte-Carlo procedures are available for
every test while the exact null distribution
is currently available for two-sample problems only.
Inference problems can be specified by a traditional formula based
interface. Support for data available as `exprSet' objects in a BioC
environment is implemented as well. The theoretical framework is sketched
in the vignette and we refer to `vignette("coin")' for further details.
Comments, suggestions, bug-reports etc. are more than welcome!
Torsten, Kurt, Mark and Achim
Title: Conditional Inference Procedures in a Permutation Test Framework
Date: $Date: 2005/06/02 14:55:45 $
Author: Torsten Hothorn and Kurt Hornik, with contributions by
Mark van de Wiel and Achim Zeileis
Maintainer: Torsten Hothorn <Torsten.Hothorn at R-project.org>
Description: Conditional inference procedures for the general independence
problem including two-sample, K-sample, correlation, censored, ordered
and multivariate problems.
Depends: R (>= 2.0.0), methods, survival, mvtnorm
More information about the R-packages