A COMPARISON OF PARAMETRIC AND NONPARAMETRIC APPROACHES TO ROC ANALYSIS OF QUANTITATIVE DIAGNOSTIC-TESTS

Receiver operating characteristic (ROG) analysis, which yields indices of accuracy such as the area under the curve (AUG), is increasingly b eing used to evaluate the performances of diagnostic tests that produc e results on continuous scales. Both parametric and nonparametric ROC approaches are available to assess the discriminant capacity of such t ests, but there are no clear guidelines as to the merits of each, part icularly with non-binormal data. investigators may worry that when dat a are non-Gaussian, estimates of diagnostic accuracy based on a binorm al model may be distorted. The authors conducted a Monte Carte simulat ion study to compare the bias and sampling variability in the estimate s of the AUCs derived from parametric and nonparametric procedures. Ea ch approach was assessed in data sets generated from various configura tions of pairs of overlapping distributions; these included the binorm al model and non-binormal pairs of distributions where one or both pai r members were mixtures of Gaussian (MG) distributions with different degrees of departures from bi-normality. The biases in the estimates o f the AUCs were found to be very small for both parametric and nonpara metric procedures. The two approaches yielded very close estimates of the AUCs and of the corresponding sampling variability even when data were generated from non-binormal models. Thus, for a wide range of dis tributions, concern about bias or imprecision of the estimates of the AUC should not be a major factor in choosing between the nonparametric and parametric approaches.