aucfn {mgrp}R Documentation

Utility function to calculate AUC of ROC

Description

This function calculates AUC of a marker. The function is not intended for general use, so will not be well documented.

Usage

aucfn(DATA, st.dev = TRUE, which.model = 1, na.rm = FALSE)

Arguments

DATA a matrix/dataframe with at least 3 columns. The first column is unused and might contain sample id information. The second column is a binary outcome (any number > 0 in this row will be treated as 1). The 3rd (and successive) column(s) contain markers. One of these columns (as selected by which.model) will be used as the marker for the output AUC.
st.dev specifies whether to return a SD estimate for the AUC
which.model selects which column of DATA to use as the predictor for the AUC estimate (default = 1 specifies the 3rd column)
na.rm specifies whether to remove rows with NAs (default=FALSE).

Value

If sd.dev== FALSE, the function returns AUC (a scalar). If sd.dev== TRUE, the function returns a 2-item dataframe with AUC & SD.

Note

This function is really not intended for general use. For more ROC related functions, please see ...

Author(s)

Daryl Morris

References

~put references to the literature/web site here ~

See Also

~~objects to See Also as help, ~~~


[Package mgrp version 1.0 Index]