. Author manuscript; available in PMC: 2011 Dec 29.

Published in final edited form as: J R Stat Soc Ser A Stat Soc. 2009 Apr;172(2):405–425. doi: 10.1111/j.1467-985X.2009.00580.x

Table 3.

Selected simulation for Cases 2 and 8 using BIC for model selection with lognormal baseline survival function

	n = 5000				n = 1000				n = 500

	M₀	M₁	M₂	M₃	M₀	M₁	M₂	M₃	M₀	M₁	M₂	M₃
Case 2: K = 1, JP=10, APC=(−5, 0)

(0.7,10)	.	99.5	0.5	.	4.6	95.2	0.2	.	47.2	52.7	0.1	.
(0.4,5)	.	99.6	0.4	.	.	99.5	0.5	.	0.1	99.1	0.8	.
(0.1,2)	.	99.3	0.7	.	.	99.5	0.5	.	.	98.6	1.4	.
(0,60)	0.9	98.5	0.6	.	29.8	70.0	0.2	.	11.1	88.8	0.1	.
(0,20)	.	99.2	0.8	.	.	99.5	0.5	.	8.3	91.0	0.7	.
(0,4)	0.0	99.7	0.3	.	0.0	99.4	0.6	.	0.0	99.6	0.4	.

Case 8: K = 2, JP=(10,15), APC=(0, −5, 0)

(0.7,10)	0.7	1.3	97.8	0.2	87.2	6.5	6.3	.	96.4	2.7	0.9	.
(0.4,5)	.	.	99.7	0.3	28.4	3.4	67.8	0.4	73.9	5.8	20.1	0.2
(0.1,2)	.	.	99.3	0.7	0.5	0.1	98.5	0.9	19.4	2.5	77.5	0.6
(0,60)	29.2	53.9	16.5	0.4	93.1	6.6	0.3	.	97.0	2.9	.	0.1
(0,20)	.	0.1	99.4	0.5	44.5	26.2	29.1	0.2	81.6	14.1	4.3	.

(0,4)	0.0	.	99.5	0.5	Calculation failed to converge