plotROC (version 1.3.3)

calculate_multi_roc: Calculate the Empirical ROC curves for multiple biomarkers

Description

Calculate empirical ROC curve estimates given labels and predictions. Designed to work with the multi_ggroc function, this takes a data.frame and computes the ROC curve for a given list of markers.

Usage

calculate_multi_roc(data, M_string, D_string)

Arguments

data

data frame containing at least 1 marker and the common class labels, coded as 0 and 1

M_string

vector of marker column names

D_string

class label column name

Value

• List of data frames containing cutoffs, and estimated true and false positive fractions

Examples

D.ex <- rbinom(100, 1, .5)
fakedata <- data.frame(M1 = rnorm(100, mean = D.ex),
   M2 = rnorm(100, mean = D.ex, sd = .4), D = D.ex)
calculate_multi_roc(fakedata, c("M1", "M2"), "D")