. Author manuscript; available in PMC: 2014 Jan 15.

Published in final edited form as: Biometrika. 2013 May 14;100(3):741–755. doi: 10.1093/biomet/ast012

Table 3.

Sensitivity analysis results for misspecified frailty distribution

CR	σ²	σ̂²	SD	β	Bias(%)	SD	SE	CP(%)
λ₀(t): Weibull type
15%	3.48	1.01	0.17	β̂₁	−1.1	.067	.08	97
15%	3.48	1.01	0.17	β̂₂	1.0	.060	.07	97
40%	3.48	1.04	0.20	β̂₁	−3.4	.086	.10	96
40%	3.48	1.04	0.20	β̂₂	3.3	.075	.09	97
15%	1.72	0.70	0.12	β̂₁	−0.9	.067	.07	96
15%	1.72	0.70	0.12	β̂₂	1.0	.059	.07	97
40%	1.72	0.72	0.15	β̂₁	−3.3	.084	.09	95
40%	1.72	0.72	0.15	β̂₂	3.2	.072	.08	96
15%	0.65	0.39	0.10	β̂₁	−0.8	.063	.07	96
15%	0.65	0.39	0.10	β̂₂	0.8	.056	.06	97
40%	0.65	0.39	0.12	β̂₁	−3.2	.080	.09	96
40%	0.65	0.39	0.12	β̂₂	3.0	.069	.08	96

λ₀ (t): reciprocal type
15%	3.48	1.04	0.16	β̂₁	−1.2	.145	.17	98
15%	3.48	1.04	0.16	β̂₂	1.4	.130	.15	97
40%	3.48	1.14	0.22	β̂₁	−5.3	.174	.19	96
40%	3.48	1.14	0.22	β̂₂	5.4	.153	.17	96
15%	1.72	0.74	0.13	β̂₁	−1.6	.141	.16	97
15%	1.72	0.74	0.13	β̂₂	1.7	.127	.14	96
40%	1.72	0.79	0.16	β̂₁	−5.4	.169	.18	95
40%	1.72	0.79	0.16	β̂₂	5.6	.151	.16	95
15%	0.65	0.38	0.09	β̂₁	−1.6	.134	.14	95
15%	0.65	0.38	0.09	β̂₂	1.8	.123	.12	94
40%	0.65	0.41	0.11	β̂₁	−6.6	.160	.18	95
40%	0.65	0.41	0.11	β̂₂	6.8	.146	.16	95

^†

CR, censoring rate; σ², true variance of frailty; σ̂;², estimate for σ²; SD, sample standard deviation; SE, mean of estimated standard errors; CP, empirical coverage probability of 95% Wald-type confidence interval.