Perform Classical Univariate ROC
Perform.UnivROC(mSetObj=NA, feat.nm, imgName, format="png", dpi=72, isAUC, isOpt, optMethod, isPartial, measure, cutoff)Input the name of the created mSetObj (see InitDataObjects)
Input the name of the feature to perform univariate ROC analysis
Input a name for the plot
Select the image format, png, of pdf.
Input the dpi. If the image format is pdf, users need not define the dpi. For png images, the default dpi is 72. It is suggested that for high-resolution images, select a dpi of 300.
Logical, select T to compute the 95 percent confidence interval band and "F" to not
Logical, show the optimal cutoff, T to show it and F to not
Select the optimal cutoff by using either closest.topleft for closest to top-left corner or youden for farthest to the diagonal line (Youden)
Logical, input T to calculate a partial ROC curve, and F to not
Select the parameter to limit the calculation of the partial ROC curve, se for the X-axis (maximum false-positive rate) and sp for the Y-axis, representing the minimum true positive-rate
Input the threshold to limit the calculation of the partial ROC curve, the number must be between 0 and 1.