The area above the ordinal dominance graph and the area below the receiver operating characteristic graph
Citations
19,398 citations
Cites background from "The area above the ordinal dominanc..."
...More important, however, was the more recent recognition by Bamber (8 ) that this " probability of correctly ranking a (normal, abnormal) pair " is intimately connected with the quantity calculated in the Wilcoxon or Mann- Whitney statistical test....
[...]
...More important, however, was the more recent recognition by Bamber (8) that this “probability of correctly ranking a (normal, abnormal) pair” is intimately connected with the quantity calculated in the Wilcoxon or MannWhitney statistical test....
[...]
...5, W is no longer nonparametric; its standard error, SE(W), depends on two distribution-specific quantities, Qi and Q2, which have the following interpretation: Q i = Prob (two randomly chosen abnormal images will both be ranked with greater suspicion than a randomly chosen normal image) Q 2 Prob (one randomly chosen abnormal image will be ranked with greater suspicion than two randomly chosen normal images) If we assume for the moment (as Green and Swets do in their proof regarding 0 and the area under the ROC curve) that the ratings are on a scale that is sufficiently continuous that it does not produce “ties,” then SE(W), or equivalently SE(area underneath empirical ROC curve), can be shown (8) to be...
[...]
16,496 citations
6,339 citations
Cites background or methods from "The area above the ordinal dominanc..."
...Analytical formulas to calculate the area are in reports by Bamber (31) and Hanley and McNeil (32)....
[...]
...(31, 32), introduced by the chemist Frank Wilcoxon....
[...]
5,651 citations
4,905 citations
References
11,820 citations
11,055 citations
3,106 citations
382 citations
234 citations