Fig. 2
From: Simulation of complex data structures for planning of studies with focus on biomarker comparison
Simulated mean ranks. The mean rank is based on 10000 simulation runs, with different methods of data generation. The dashed line represents the reference mean ranks from population data. Normal data: multivariate normal distributed covariate data. Quantile data: covariate data drawn from the empirical distribution. GLM: modeling of the relationship with logistic regression. GAM: modeling of the relationship with generalized additive models