Table 3.

Comparison of model accuracy and stability using simulation data (95% CIs of the area under the curve [AUC] for hold-out validation)^a.

Study model	Accuracy ≥0.80 (training sets)				Stability ≥0.70 (testing sets)
	Training cases, N	AUC (95% CI)	Significant difference	Testing cases, N		AUC (95% CI)	Significant difference
Naïve Bayes (1)	700	0.90 (0.88-0.92)	1, 2	300		0.89 (0.85-0.93)	—b
KNNc (2)	700	0.99 (0.98-1.00)	1, 3	300		0.91 (0.88-0.94)	3
LRd (3)	700	0.87 (0.85-0.89)	1, 2	300		0.81 (0.77-0.85)	2

^aThe computation of the 95% CI for the AUC is referred to in equations (10) to (12).

^bData not available.

^cKNN: k-nearest neighbors.

^dLR: logistic regression.