Table 2.

Comparison of discrimination performance of optimal prediction models.

Predictive model	AUC difference	P-Value	cNRI	P-Value	IDI	P-Value
Lasso-AdaBoost vs. FLR-L1-LR	0.019	0.002	0.208 (0.078, 0.337)	<0.001	0.032 (0.019, 0.045)	<0.010
Lasso-AdaBoost vs. FLR-RF	0.007	0.334	0.097 (−0.033, 0.228)	0.143	0.016 (0.007, 0.025)	<0.010
Lasso-AdaBoost vs. FLR-SVM	0.016	0.047	0.167 (0.037, 0.296)	0.012	0.029 (0.016, 0.042)	<0.010
FLR-RF vs. FLR-L1-LR	0.012	0.045	0.108 (−0.022, 0.238)	0.105	0.016 (0.003, 0.028)	0.010
FLR-RF vs. FLR-SVM	0.003	0.016	0.072 (−0.058, 0.203)	0.278	0.013 (0.001, 0.026)	0.040
FLR-SVM vs. FLR-L1-LR	0.010	0.118	0.278 (0.149, 0.408)	<0.001	0.003 (0.001, 0.004)	<0.010

AUC, area under the receiver operating characteristic curve; cNRI, continuous Net Reclassification Index; IDI, Integrated Discrimination Improvement Index; Lasso-AdaBoost, AdaBoost with Lasso regression; FLR-L1-LR, L1 regularized Logistic regression with forward Partial Likelihood Estimation; FLR-RF, random forest with forward Partial Likelihood Estimation; FLR-SVM, support vector machine with forward Partial Likelihood Estimation.