auc.cph

auc.coxph

<p>can be logical or characters. FALSE means no model TP and FP,
characters mean model names.</p>

model

<p>can be logical or characters. TRUE means all x variable in regression
will be calculated. One or more characters will be calculated only.</p>

method

The ROC curve method is one of the most important and commonly used
methods for model accuracy assessment, which is one of the most important elements
of model evaluation.
The 'modelROC' package is a model-based ROC assessment tool, which directly works
for ROC analysis of regression results for logistic regression of binary variables,
including the glm() and lrm() commands, and COX regression for survival analysis,
including the cph() and coxph() commands.
The most important feature of 'modelROC' is that both the model and the independent
variables can be analysed simultaneously, and for survival analysis
multiple time points and area under the curve analysis are supported.
Still, flexible visualisation is possible with the 'ggplot2' package.
Reference are Kelly H. Zou (1998) <doi:10.1002/(sici)1097-0258(19971015)16:19%3C2143::aid-sim655%3E3.0.co;2-3> and
P J Heagerty (2000) <doi:10.1111/j.0006-341x.2000.00337.x>.

auc: auc for model

Description

Usage

Arguments

Value