Fig. 5
From: ecpc: an R-package for generic co-data models for high-dimensional prediction
Simulation study for variable selection based on 50 training and test sets for various types of co-data. a) Average sensitivity and precision for several methods and various tuning parameters; b) Mean squared error prediction performance on the test data. The lines indicate the pointwise average and the inner and outer shaded bands indicate the 25–75% and 5-95% quantiles respectively