Skip to main content
. Author manuscript; available in PMC: 2019 Aug 20.
Published in final edited form as: J Comput Graph Stat. 2018 Aug 20;27(4):763–772. doi: 10.1080/10618600.2018.1474115

Figure 1. Variable importance from modified Random Survival Forests.

Figure 1

Barplot of variable importance for each of 100 covariates, considering the setting in which 1 − SJ+1 = 0.10, φ1 = 0.61, φ0 = 0.995 and there are no missed visits. (a) Continuous covariates - average variable importance over 100 simulated datasets is shown, where the first five (shown in red) represent the true biomarkers. (b) GWAS data - variable importance for a representative simulation is shown (results are not averaged as the identity of the true biomarkers varies between simulations). ’Blue’ indicates true biomarkers with MAF ∈ (0.35, 0.5] and ‘red’ indicates true biomarkers with MAF ∈ (0, 0.35]. (c) Omics data - average variable importance over 100 simulated datasets is shown, where the first five (shown in red) represent the true biomarkers.