Fig. 3

Model performance in the internal validation setting, measured by the integrated calibration index (ICI), Brier score, and area under the receiver operating curve (AUC). Cognitive impairment was predicted using logistic regression, lasso regression, random forest (rf), and generalized boosted regression (gbm) prediction models. Models were trained either with all predictor variables, only parent nodes (direct causes) of the outcome, only children nodes (consequences) of the outcome, or with the exogenous variables age, sex, and APOE ε4 (apoe4). Depicted are the full distributions of ICI, Brier scores, and AUC, smoothed with a Gaussian kernel density function and medians marked with ◊. The displayed metrics were obtained from 10,000 repetitions of data generation and model training on the first imputed dataset