Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Allowing for mandatory covariates in boosting estimation of sparse high-dimensional survival models

Figure 3

Variability of the .632+ prediction error estimates. Integrated prediction error curve estimates from single bootstrap samples for CoxPath, gradient boosting, componentwise CoxBoost, and an IPI-only Cox model ("IPI"), for corresponding models where the IPI is included as a an additional optional covariate ("...+IPI"), and for CoxBoost fits that incorporate the IPI as a mandatory covariate ("IPI mandatory"). The Kaplan-Meier benchmark value is indicate by a horizontal line.

Back to article page