harmonicmeanp: Harmonic Mean p-Values and Model Averaging by Mean Maximum Likelihood

The harmonic mean p-value (HMP) test combines p-values and corrects for multiple testing while controlling the family-wise error rate in a way that is more powerful than common alternatives including Bonferroni and Simes procedures, more stringent than controlling the false discovery rate, and is robust to positive correlations between tests and unequal weights. It is a multi-level test in the sense that a superset of one or more significant tests is almost certain to be significant and conversely when the superset is non-significant, the constituent tests are almost certain to be non-significant. It is based on MAMML (model averaging by mean maximum likelihood), a frequentist analogue to Bayesian model averaging, and is theoretically grounded in generalized central limit theorem. For detailed examples type vignette("harmonicmeanp") after installation.

Version: 1.1
Depends: FMStable, knitr, ape
Published: 2019-01-07
Author: Daniel J. Wilson
Maintainer: Daniel Wilson <hmp.R.package at gmail.com>
License: GPL-3
NeedsCompilation: no
CRAN checks: harmonicmeanp results


Reference manual: harmonicmeanp.pdf
Vignettes: harmonicmeanp
Package source: harmonicmeanp_1.1.tar.gz
Windows binaries: r-devel: harmonicmeanp_1.1.zip, r-release: harmonicmeanp_1.1.zip, r-oldrel: harmonicmeanp_1.1.zip
OS X binaries: r-release: harmonicmeanp_1.1.tgz, r-oldrel: harmonicmeanp_1.1.tgz
Old sources: harmonicmeanp archive

Reverse dependencies:

Reverse suggests: treedater


Please use the canonical form https://CRAN.R-project.org/package=harmonicmeanp to link to this page.