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 P., Quantifying performance of a diagnostic test as the expected information for discrimination: relation to the C-statistic. Statistical Methods for Medical Research 2018, in press). 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.

Downloads:

Linking:

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