eNetXplorer (version 1.1.2)

plotSurvROC: generates time-dependent ROC plots from Cox predicted risks

Description

For Cox regression models, this function generates time-dependent ROC plot(s) (true positive rate vs false positive rate) for a given alpha at the timepoint(s) provided based on median predicted risk. Provided that the eNetXplorer object was generated with survAUC=T, the cross-validated median AUC and 95% CI are shown in the default title. For more details, see Heagerty et al and package survivalROC.

Usage

plotSurvROC(x, alpha.index=NULL, survAUC_time,
xlab="False positive rate (1 - Specificity)",
ylab="True positive rate (Sensitivity)", cex.lab=1, main=NULL, col.main="black", 
cex.main=0.95, status0="censored", status1="events", …)

Arguments

x

eNetXplorer object (must be family="cox").

alpha.index

Integer indices to select alpha values. Default is 1:length(alpha)

survAUC_time

Timepoint(s) of interest. Must be in the same time units as the survival time provided to build the eNetXplorer object.

xlab

Custom x-axis label.

ylab

Custom y-axis label.

cex.lab

Axis label size.

main

Custom title.

col.main

Title color.

cex.main

Title size.

status0

Title label for censoring ("status"=0).

status1

Title label for events ("status"=1).

…

Additional parameters.

References

Blanche P, Dartigues J-F and Jacqmin-Gadda H. Estimating and comparing time-dependent areas under receiver operating characteristic curves for censored event times with competing risks, Statistics in Medicine (2013) 32:5381-5397.

See Also

eNetXplorer, plot

Examples

# NOT RUN {
data(breastCancerSurv)
fit = eNetXplorer(x=breastCancerSurv$predictor, y=breastCancerSurv$response, family="cox", 
n_run=25, n_perm_null=15, seed=111, survAUC=TRUE, survAUC_time=c(1,5)*365)
plot(x=fit, plot.type="survROC", survAUC_time=c(1,5)*365, status0="censored", status1="deaths")
# }