bouldr: Run ROCs and pairwise comparisons using desired test

In jlangaa/bouldr: A Psychology-Focused Tool for ROC Analyses

Run ROCs and pairwise comparisons using desired test

Description

This function takes a data frame and a formula and runs nested ROCs as appropriate. Also required are the levels of the response variable and the direction of the cases (i.e., which level is the positive case). Providing an argument for test will change the test used to compare ROC curves.

Usage

bouldr(formula, data, levels, direction = "auto", test = "delong", ...)

Arguments

formula	A specification of the variables of form: response ~ predictor + (group1) + (group2)
data	A data frame containing the response and predictor vars
levels	The levels of the response variable as a character vector (e.g., C("yes","no"))
direction	Either "<" or ">", indicating which level is the positive case
test	Which test should be used to compare ROCs (optional). Default is delong
...	Additional arguments

Value

An object of class 'bouldr' containing ROCs and tests.

Examples

dd <- generate_data(10,Diagnosis = c("Depression","Anxiety","ADHD"),
        Measure = c("A","B"), Informant = c("Self","Parent", "Teacher"))
single <- bouldr(dat = dd,
f = Outcome ~ Score,
test = 'delong',
levels = c('no','yes'),
direction = "<")
aucs(single)

jlangaa/bouldr documentation built on May 3, 2024, 5:40 a.m.