Empirical evaluation of internal validation methods for prediction in large-scale clinical data with rare-event outcomes: a case study in suicide risk prediction

Table 2 Estimated AUC of prediction models from split-sample and entire sample estimation approaches in the development dataset and prospective validation dataset

Prediction model estimation	Internal validation approach	AUC (95% CI)
Prediction model estimation	Internal validation approach	Development dataset	Prospective dataset
Split-sample	Validate in testing set	0.846 (0.817, 0.870)	0.814 (0.771, 0.851)
Entire-sample	Cross-validation	0.832 (0.812, 0.851)	0.811 (0.768, 0.849)
Entire-sample	Bootstrap optimism correction	0.878 (0.861, 0.890)	0.811 (0.768, 0.849)

ISSN: 1471-2288