Conducting the non-inferiority test for the means with unknown coefficient of variation in a three-arm trial

Table 1 The type I error rates for testing non-inferiority with non-inferiority limit = 0.8 in \(\tau_{R} = 1\), \(\mu_{R} - \mu_{P} =\) 9, 15 and 20, respectively

\(\mu_{R} - \mu_{P}\)	\(\tau_{P}\)	\(n\)	GP	DM	EB
9	0.5	60	0.0497	0.0001	0.0001
		90	0.0503	0.0001	0.0003
		120	0.0506	0.0002	0.0009
		480	0.0511	0.0016	0.0082
		900	0.0518	0.0038	0.0264
	1.0	60	0.0486	0.0001	0.0015
		90	0.0493	0.0002	0.0018
		120	0.0498	0.0002	0.0045
		480	0.0501	0.0003	0.0318
		900	0.0506	0.0004	0.0436
	2.0	60	0.0488	0.0001	0.0091
		90	0.0499	0.0001	0.0127
		120	0.0504	0.0002	0.0264
		480	0.0506	0.0002	0.0382
		900	0.0511	0.0003	0.0436
15	0.5	60	0.0475	0.0008	0.0018
		90	0.0481	0.0009	0.0055
		120	0.0482	0.0019	0.0091
		480	0.0499	0.0036	0.0273
		900	0.0502	0.0043	0.0364
	1.0	60	0.0493	0.0004	0.0155
		90	0.0501	0.0004	0.0191
		120	0.0505	0.0007	0.0218
		480	0.0508	0.0007	0.0423
		900	0.0514	0.0015	0.0445
	2.0	60	0.0500	0.0001	0.0327
		90	0.0503	0.0002	0.0364
		120	0.0504	0.0002	0.0373
		480	0.0507	0.0003	0.0455
		900	0.0512	0.0004	0.0473
20	0.5	60	0.0495	0.0021	0.0073
		90	0.0503	0.0025	0.0127
		120	0.0511	0.0032	0.0164
		480	0.0514	0.0047	0.0355
		900	0.0516	0.0058	0.0464
	1.0	60	0.0487	0.0006	0.0264
		90	0.0493	0.0010	0.0291
		120	0.0504	0.0010	0.0303
		480	0.0510	0.0012	0.0455
		900	0.0513	0.0015	0.0467
	2.0	60	0.0486	0.0001	0.0418
		90	0.0490	0.0001	0.0436
		120	0.0502	0.0002	0.0445
		480	0.0509	0.0003	0.0464
		900	0.0515	0.0005	0.0477

\(n\) The total sample sizes, GP The GPV-based method, DM The Delta method and EB The empirical bootstrap method

ISSN: 1471-2288