Combining techniques for screening and evaluating interaction terms on high-dimensional time-to-event data

BMC Bioinformatics

Table 3 Results of the simulation study for rsf-VIF-res in scenarios Sim22_0.25 - Sim22_2.5 and Sim_bin

Scenario	IntScreen	IntSensiA	VarsTotal	MainSensi	IntSensi	rIPEC
						CoxBoostM	Final Model
Strategy rsf-VIF-res
Sim22_2.5	15231.62	0.3 (0.05)	16.04 (14.72)	0.29 (0.05)	0.3 (0.05)	0 (0.06)	0.1 (0.16)
Sim22_2.0	17200.98	0.33 (0.05)	18.28 (14.41)	0.51 (0.05)	0.33 (0.05)	0.01 (0.06)	0.14 (0.19)
Sim22_1.5	16878.62	0.34 (0.05)	20.38 (14.57)	0.67 (0.05)	0.34 (0.05)	0.05 (0.08)	0.12 (0.15)
Sim22_1.0	17701.72	0.4 (0.05)	19.9 (12.93)	0.81 (0.04)	0.39 (0.05)	0.12 (0.09)	0.14 (0.14)
Sim22_0.5	18613.08	0.43 (0.05)	18.04 (10.92)	0.93 (0.03)	0.13 (0.03)	0.21 (0.1)	0.17 (0.1)
Sim22_0.25	20404.02	0.46 (0.05)	20.4 (11.63)	0.98 (0.01)	0 (0)	0.25 (0.11)	0.2 (0.11)
Sim22_bin	22847.06	0.42 (0.05)	26.22 (10.9)	0.49(0.07)	0.34(0.05)	0.2 (0.07)	0.15 (0.08)

IntScreen (given as mean) is the number of selected interactions by the corresponding screening method; IntSensiA (given as ‘sensitivity value (sd)’) is the sensitivity related to the availability of true interactions; VarsTotal (given as ‘mean (sd)’) is the number of total variables in the final model; MainSensi (given as ‘sensitivity value (sd)’) is the sensitivity related to the inclusion of true main effects; and IntSensi (given as ‘sensitivity value (sd)’) is the sensitivity related to the inclusion of true interactions. The rIPEC values (given as ‘mean (sd)’) are shown for CoxBoostM and the final model. The scenarios were repeated 50 times. Additional file 1: Figure S2 provides boxplots for further insights into the nature of the variability in rIPEC.

ISSN: 1471-2105