Exploring predictors of welfare dependency 1, 3, and 5 years after mental health-related absence in danish municipalities between 2010 and 2012 using flexible machine learning modelling

Table 5 Validation results

Follow-up	1 year				3 years				5 years
Model	MLR	PMLR_im	XGB	XGB_im	MLR	PMLR_im	XGB	XGB_im	MLR	PMLR_im	XGB	XGB_im
AUC-ROC
Apparent	0.71	0.71	0.73	0.72	0.70	0.70	0.72	0.72	0.71	0.70	0.72	0.72
tenfold-cv	0.71	0.71	0.71	0.71	0.69	0.70	0.70	0.69	0.70	0.70	0.70	0.70
Test set	0.71	0.70	0.71	0.71	0.70	0.70	0.71	0.71	0.70	0.70	0.71	0.70
Brier-scaled
Apparent	0.13	0.13	0.15	0.15	0.10	0.10	0.12	0.12	0.10	0.08	0.10	0.12
tenfold-cv	0.11	0.11	0.13	0.13	0.09	0.09	0.10	0.10	0.10	0.10	0.10	0.10
Test set	0.12	0.12	0.14	0.14	0.12	0.12	0.13	0.12	0.12	0.12	0.10	0.12

MLR Multinomial logistic regression, PMLR _im Penalized multinomial logistic regression (ridge penalty) with single imputation, XGB XGBoost, XGB _im XGBoost with single imputation

ISSN: 1471-2458