BMC Medical Research Methodology

Table 6 Results of the association parameter α obtained from joint model and TVCM fitted to data generated considering a non-monotonic baseline hazard function (scenario 9 of Table 1), α∈(0.3,0.6) and σ_ε∈(0.1,0.3,0.5) with CV ∈(3.1%,9.4%,15.6%)

From: Joint model robustness compared with the time-varying covariate Cox model to evaluate the association between a longitudinal marker and a time-to-event endpoint

		α=0.3					α=0.6
σ_ε	Model	Est	ESE	ASE	%Bias	CP	Est	ESE	ASE	%Bias	CP
0.1	TVCM(1x/week)	0.292	0.058	0.058	-2.7	94	0.579	0.062	0.061	-3.5	93
	joint-constant	0.183	0.041	0.054	-39	39	0.336	0.035	0.052	-44	0
	joint-weibull	0.337	0.066	0.058	12	86	0.638	0.065	0.061	6.3	89
	joint-spline	0.303	0.060	0.059	1.0	95	0.608	0.064	0.063	1.3	94
0.3	TVCM(1x/week)	0.274	0.056	0.056	-8.7	92	0.538	0.059	0.058	-10.3	80
	joint-constant	0.180	0.043	0.055	-40	39	0.328	0.037	0.053	-45	0
	joint-weibull	0.340	0.068	0.059	13	87	0.642	0.070	0.065	7.0	89
	joint-spline	0.304	0.062	0.061	1.3	95	0.608	0.077	0.067	1.3	94
0.5	TVCM(1x/week)	0.244	0.053	0.053	-18	81	0.471	0.055	0.055	-21	36
	joint-constant	0.174	0.046	0.052	-42	39	0.312	0.04	0.055	-48	0
	joint-weibull	0.344	0.071	0.062	15	85	0.643	0.164	0.071	7.2	89
	joint-spline	0.304	0.065	0.064	1.3	95	0.615	0.197	0.073	2.5	94

Mean of the maximum likelihood estimates (Est), empirical Monte Carlo standard error (ESE), asymptotic standard error (ASE), percentage bias (%Bias) and 95% coverage probabilities (CP) are shown

Back to article page

ISSN: 1471-2288

Contact us

General enquiries: journalsubmissions@springernature.com