Daim (version 1.1.0)

deLong.test: DeLong test

Description

A nonparametric test for comparing AUC of two or more correlated ROC curves.

Usage

deLong.test(x, labels, labpos, ref=NULL, conf.level=0.95)

Arguments

x

at least 2 dimensional matrix containing the observations of numeric predictors.

labels

a vector containing the true class labels. It can be a factor or character vector.

labpos

a character string of the variable labels that defines a "positive" event.

ref

an index of reference predictors. Relevant only by pairwise comparisons.

conf.level

confidence level of the interval.

References

DeLong, E.R., DeLong, D.M., and Clarke-Pearson, D.L. (1988). Comparing the Areas Under Two or More Correlated Receiver Operating Characteristics Curves: A Nonparametric Approach. Biometrics, 44, 837--845.

See Also

roc, Daim, plot.Daim, performDaim

Examples

y <- rbinom(100, 1, .5)
x1 <- rnorm(100) + 1.5 * y
x2 <- rnorm(100) + .5 * y
x3 <- rnorm(100) + 2.5 * y
x <- data.frame(x1,x2,x3)
y <- as.factor(y)

(M1 <- deLong.test(x, labels=y, labpos="1"))



## Not run:   
# ## just because of checking time on CRAN
#   
# 
# data(iris)
# (M2 <- deLong.test(iris[,1:4], labels=iris[,5], labpos="versicolor"))
# ## End(Not run)