CopulaDTA: Copula Based Bivariate Beta-Binomial Model for Diagnostic Test Accuracy Studies

Modelling of sensitivity and specificity on their natural scale using copula based bivariate beta-binomial distribution to yield marginal mean sensitivity and specificity. The intrinsic negative correlation between sensitivity and specificity is modelled using a copula function. A forest plot can be obtained for categorical covariates or for the model with intercept only. Nyaga VN, Arbyn M, Aerts M (2017) <doi:10.18637/jss.v082.c01>.

Version: 1.0.0
Depends: R (≥ 3.4.2), rstan (≥ 2.16.2)
Imports: methods, ggplot2 (≥ 2.2.1), plyr (≥ 1.8.4), stats (≥ 3.4.2), reshape2 (≥ 1.4.2), grDevices (≥ 3.4.2)
Suggests: knitr, loo, Rmisc, httr, bayesplot
Published: 2017-11-23
Author: Victoria N Nyaga [aut, cre]
Maintainer: Victoria N Nyaga <victoria.nyaga at outlook.com>
License: GPL-2
NeedsCompilation: no
Citation: CopulaDTA citation info
CRAN checks: CopulaDTA results

Downloads:

Reference manual: CopulaDTA.pdf
Vignettes: CopulaDTA Package Vignette
Package source: CopulaDTA_1.0.0.tar.gz
Windows binaries: r-devel: CopulaDTA_1.0.0.zip, r-release: CopulaDTA_1.0.0.zip, r-oldrel: CopulaDTA_0.0.5.zip
OS X El Capitan binaries: r-release: CopulaDTA_1.0.0.tgz
OS X Mavericks binaries: r-oldrel: CopulaDTA_0.0.5.tgz
Old sources: CopulaDTA archive

Linking:

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