Table 4.

Simulation results under additive MRL models

N	β	censoring rate		Full†		CC (30%)‡		NCC (1-to-l)§
				β1	β2	β1	β1	β2	β1
5000	(0,0)	0.7	Bias	0.000	0.000	0.000	−0.002	0.001	−0.001
			SD	0.020	0.035	0.025	0.044	0.025	0.042
			SE	0.020	0.034	0.025	0.044	0.024	0.041
			CP	94.6	94.2	95.9	94.9	93.4	94.2
5000	(0,0)	0.8	Bias	0.000	0.001	−0.001	0.000	−0.001	0.002
			SD	0.026	0.047	0.032	0.059	0.032	0.060
			SE	0.026	0.045	0.033	0.057	0.033	0.057
			CP	94.8	94.5	95.3	94.5	95.2	93.5
5000	(0,0)	0.9	Bias	0.000	0.004	0.002	0.001	0.001	0.005
			SD	0.039	0.068	0.045	0.074	0.052	0.091
			SE	0.038	0.066	0.044	0.076	0.050	0.088
			CP	94.8	94.7	94.4	95.7	94.6	94.2
5000	(0.2,0.2)	0.7	Bias	0.004	0.005	0.005	0.004	0.005	0.006
			SD	0.029	0.050	0.038	0.066	0.035	0.059
			SE	0.029	0.049	0.037	0.063	0.033	0.058
			CP	95.0	93.7	95.0	94.4	93.9	94.7
5000	(0.2,0.2)	0.8	Bias	0.004	0.006	0.009	0.001	0.005	0.008
			SD	0.039	0.067	0.051	0.083	0.048	0.084
			SE	0.039	0.065	0.049	0.082	0.048	0.082
			CP	94.7	94.1	95.0	94.8	94.8	94.2
5000	(0.2,0.2)	0.9	Bias	0.000	0.006	−0.001	0.000	−0.003	0.002
			SD	0.063	0.105	0.074	0.121	0.082	0.135
			SE	0.058	0.098	0.067	0.114	0.079	0.136
			CP	94.1	94.0	93.0	93.7	94.1	95.4

SD: sample standard deviation; SE: mean of estimated standard error;

CP: empirical coverage probability of 95% confidence interval;

^†Full-cohort population

^‡Case-Cohort design with 30% random sample from full-cohort population

^§Nested Case-Control design with 1 control for each case