Bugs are always possible, I make my best to provide a new release averytime I found them. Please report bugs to: ‘livio.finos@unipd.it’. Peculiarities, feedback and general queries on the use and theory of the software are also welcomed.

A R package for multivariate permutation tests.

The github dev version can be found here.

Some example here https://livioivil.github.io/flipscores/

J Hemerik, JJ Goeman and L Finos (2019) Robust testing in generalized
linear models by sign-flipping score contributions. Journal of the Royal
Statistical Society Series B: Statistical Methodology, Volume 82, Issue
3, July 2020, Pages 841–864.

https://doi.org/10.1111/rssb.12369

R De Santis, J Goeman, J Hemerik, L Finos (2022) Inference in
generalized linear models with robustness to misspecified variances
arXiv: 2209.13918.

https://arxiv.org/abs/2209.13918

A R package for multivariate permutation tests.

**flipscores**
A R package for robust score test with permutation tests.

<img title=“permSpace.png” alt=“permSpace.png” src=“../images/permSpace.png” mce_real_src=“../images/permSpace.png”

**DESCRIPTION**: NPClib is the acronyms of “NonParametric
Combination library”, the guiding principle underlying this software
package. It is a library of MATLAB functions and it allows for a fully
univariate and multivariate permutation-based inference. Most of the
methods are formally stated in Pesarin, F. (2001) Multivariate
Permutation Test with Application in Biostatistics. Wiley, New York.
Other can be found in

Pesarin and Salmaso (2010)*Permutation Tests for Complex Data -
Theory, Applications and Software*, Wiley Series in Probability and
Statistics.

It’s a (small) collection of functions for Multiplicty Correction and
Multiple Testing. Among others, you can find functions from some of my
work

**lsd.test()DESCRIPTION**: It performs test on
multivariate-multiple linear models (i.e. many predictors and predicted)
with covariates (i.e. not under test). It also allows to deal with
datasets having more dependent variables than observations (negative
df).

See L. Finos (2011). A note on Left-Spherically Distributed Test with covariates, Statistics and Probabilty Letters, accepted.

**step.adj()DESCRIPTION**: This function corrects the false
discoveries (weak familywise control) due to model selection. It works
with models of class glm and selected with function step() of library
stats.

See L. Finos, C. Brombin, L. Salmaso (2009). Adjusting stepwise p-values in generalized linear models. Communications in Statistics - Theory and Methods. Accepted.

and L. Finos, C. Brombin, L. Salmaso (2007). Adjusting p-values of a stepwise generalized linear model. MCP 2007 Vienna, Austria. July 8-11, 2007. slides

C. Brombin, L. Finos, R. Arboretti Giancristofaro (2007). How confident can you be in stepwise-glm findings? First IBS Channel Network meeting, Rolduc, May 8-11, 2007.

**p.adjust.w()DESCRIPTION**: Given a set of p-values, returns
p-values adjusted using one of several (weighted) methods. It extends
the method of p.adjust of library stats.

See L. Finos, L. Salmaso (2007). FDR- and FWE-controlling methods
using data-driven weights. Journal of Statistical Planning and
Inference, 137,12, 3859-3870. ISSN: 0378-3758. See also the related
research
page

**DESCRIPTION**: This library collects some procedures controlling
the Generalized Familywise Error Rate: kfwe.ord ordinal, Lehamn and
Romano, Guo and Romano (Single Step and Step Down) procedures.

More datails here

**AUTHORS:** Livio Finos and Alessio Farcomeni

**DESCRIPTION**: MTHw perform Multiple Hypotheses Testing (FWE and
FDR) procedures with weights as described in: Benjamini,Y., Hochberg,Y.,
1997. Multiple hypotheses testing with weights. Scand. J. Statist. 24,
407-418.

The weights must usually be chosen a priori, on the basis of experimental hypotheses. Under some conditions, however, they can be chosen making use of information from the data (therefore a posteriori) while maintaining multiplicity control.

More datails here

**FILES**: FDRw.m, BHw.m, HOMw.m