flip (version 2.0)

flip-package: The library is devoted to permutation-based inferential methods.

Description

It implements many univariate and multivariate permutation (and rotation) tests.

The tests comprised are: the one and two samples, ANOVA, linear models, Chi Squared test, rank tests (i.e. Wilcoxon, Mann-Whitney, Kruskal-Wallis), Kolmogorov-Smirnov and Anderson-Darling.

Test on Linear Models are performed also in presence of covariates (i.e. nuisance parameters).

The permutation and the rotation method to get the null distribution of the test statistic(s) are available.

It also implements methods for multiplicity control such as Westfall-Young min-p procedure and Closed Testing (Marcus, 1976).

Details

ll{ Package: flip Type: Package Version: 1.1 Date: 2012-02-05 License: GPL

References

Pesarin (2001) Multivariate Permutation Tests with Applications in Biostatistics. Wiley, Mew York.

I widely borrowed the structure of flip from library globaltest of (http://www.bioconductor.org/packages/release/bioc/html/globaltest.html, Authors: J. Goeman and J. Oosting).

Examples

Y=data.frame(matrix(rnorm(50),10,5))
names(Y)=LETTERS[1:5]
Y[,1:2]=Y[,1:2]
x=rep(0:1,5)
data=data.frame(x=x, Z=rnorm(10))
res = flip(Y+matrix(x*2,10,5),~x,~Z,data=data)
res

plot(res)

p2=npc(res,"fisher",subsets=list(c1=c("A","B"),c2=names(Y)))
p2