se_auc {auctestr} | R Documentation |
Compute standard error of AUC score, using its equivalence to the Wilcoxon statistic.
se_auc(auc, n_p, n_n)
auc |
value of A' statistic (or AUC, or Area Under the Receiver operating characteristic curve) (numeric). |
n_p |
number of positive cases (integer). |
n_n |
number of negative cases (integer). |
Hanley and McNeil, The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology (1982) 43 (1) pp. 29-36.
Fogarty, Baker and Hudson, Case Studies in the use of ROC Curve Analysis for Sensor-Based Estimates in Human Computer Interaction, Proceedings of Graphics Interface (2005) pp. 129-136.
Other fbh method: auc_compare
,
fbh_test
se_auc(0.75, 20, 200)
## standard error decreases when data become more balanced over
## positive/negative outcome class, holding sample size fixed
se_auc(0.75, 110, 110)
## standard error increases when sample size shrinks
se_auc(0.75, 20, 20)