Hi, I am working on a ROC plot problem. Thanks in advance for any help. We run 20 cross validation (CV) on our data for each customized threshold, i.e. from 0.1 to 0.9. Then for these 9 threshold values, we get the overall performance, that is we are able to get overall sensitivity and specificity across each 20CVs using different threshold. So we have in total 9 points to be plotted on a ROC graph. But the question is it seems that the ROC functions only read the prediction label and the actual label for each case, instead of the SEC and the SPE values. So we are looking for a way to feed the SEC and SPC values as input in order to plot ROC. Thanks. -- View this message in context: http://r.789695.n4.nabble.com/ROC-plot-with-SEC-and-SPC-as-input-tp3652562p3652562.html Sent from the R help mailing list archive at Nabble.com.