Bias-Corrected Methods for Estimating the ROC Surface of Continuous Diagnostic Tests

The bias-corrected estimation methods for the receiver operating characteristics ROC surface and the volume under ROC surfaces (VUS) under missing at random (MAR) assumption.


Version 1.0-3 (16th Oct 2018)

  • Implement Jackknife approach for estimating standard error of FULL estimate of VUS
  • Fix the bug of bootstrap estimation for variance
  • Add argument "trace" into function "vus()" for tracing information on the progress of the estimation
  • Corrected the help file of package

Version 1.0-2 (3rd July 2017)

  • Fixed the Makevars and
  • Corrected the hypothesis testing for VUS

Version 1.0-1 (17th November 2016)

  • First release.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.0-5 by Duc-Khanh To, 5 months ago

Browse source code at

Authors: Duc-Khanh To , with contributions from Monica Chiogna and Gianfranco Adimari

Documentation:   PDF Manual  

GPL (>= 2) license

Imports Rcpp

Depends on nnet, rgl, boot, stats, utils, graphics, parallel

Suggests webshot2, knitr, rmarkdown, markdown

Linking to Rcpp, RcppArmadillo

See at CRAN