wevid: Quantifying Performance of a Binary Classifier Through Weight of Evidence

The distributions of the weight of evidence (log Bayes factor) favouring case over noncase status in a test dataset (or test folds generated by cross-validation) can be used to quantify the performance of a diagnostic test (McKeigue (2018), <doi:10.1177/0962280218776989>). The package can be used with any test dataset on which you have observed case-control status and have computed prior and posterior probabilities of case status using a model learned on a training dataset. To quantify how the predictor will behave as a risk stratifier, the quantiles of the distributions of weight of evidence in cases and controls can be calculated and plotted.

Version: 0.6.1
Depends: R (≥ 2.10)
Imports: ggplot2, mclust, pROC (≥ 1.9), reshape2, zoo
Published: 2019-05-13
Author: Paul McKeigue ORCID iD [aut], Marco Colombo ORCID iD [ctb, cre]
Maintainer: Marco Colombo <mar.colombo13 at gmail.com>
License: GPL-3
URL: http://www.homepages.ed.ac.uk/pmckeigu/preprints/classify/wevidtutorial.html
NeedsCompilation: no
Citation: wevid citation info
CRAN checks: wevid results


Reference manual: wevid.pdf
Package source: wevid_0.6.1.tar.gz
Windows binaries: r-devel: wevid_0.6.1.zip, r-release: wevid_0.6.1.zip, r-oldrel: wevid_0.6.1.zip
OS X binaries: r-release: wevid_0.6.1.tgz, r-oldrel: wevid_0.6.1.tgz
Old sources: wevid archive


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