Joint Modelling Approaches to Survival Analysis via Likelihood-Based Boosting Techniques

Table 1

Shrinkage and variable selection properties regarding longitudinal and survival outcomes averaged over 100 simulation runs of the low-dimensional scenario.

	(sd)	(sd)	(sd)	(sd)	TP	FDR
True	2	1	2	1
JM	1.998	0.994	2.008	1.002	—	—	—
	(0.03)	(0.07)	(0.07)	(0.07)
lbbJM^a	1.760	0.914	1.922	0.923	1.00	0.23	108.25
	(0.08)	(0.07)	(0.07)	(0.07)
lbbJM^b	1.992	0.994	2.008	1.002	—	—	—
	(0.03)	(0.07)	(0.07)	(0.07)
	(sd)	(sd)	(sd)	(sd)	TP	FDR
True	0.5	1	2	-2
JM	0.457	0.903	1.807	-1.800	—	—	—
	(0.04)	(0.08)	(0.12)	(0.12)
lbbJM^a	0.390	0.728	1.521	-1.516	1.00	0.27	209.2
	(0.03)	(0.06)	(0.07)	(0.07)
lbbJM^b	0.373	0.713	1.500	-1.495	1.00	0.22	196.9
	(0.03)	(0.06)	(0.07)	(0.08)
glmnet	0.427	0.909	1.833	-1.823	1.00	0.51	—
	(0.03)	(0.07)	(0.11)	(0.10)