Table 5.

REML estimation of thrombolytic drugs data: Wald χ² statistics testing consistency, selected estimated treatment effects and their standard errors, comparing standard approach with data augmentation approach for various choices of h and m and the reference treatment

		Data augmentation approach
	Standard approach	Reference treatment	m	0.08			0.5
			h	0.001	0.01	0.1	0.001	0.01	0.1
Wald test of consistency	8.61	A		8.61	8.61	8.60	8.60	8.59	8.42
		B		8.61	8.61	8.62	8.61	8.62	8.64
		C		8.61	8.61	8.60	8.60	8.59	8.35
Treatment effect δAB	−0.161	A		−0.161	−0.161	−0.161	−0.161	−0.161	−0.162
		B		−0.161	−0.161	−0.161	−0.161	−0.161	−0.158
		C		−0.161	−0.161	−0.161	−0.161	−0.161	−0.162
Treatment effect δAG	−0.197	A		−0.197	−0.197	−0.197	−0.197	−0.198	−0.205
		B		−0.197	−0.197	−0.198	−0.197	−0.198	−0.202
		C		−0.197	−0.197	−0.197	−0.197	−0.198	−0.202
Standard error se(δAB)	0.046	A		0.046	0.046	0.046	0.046	0.046	0.046
		B		0.046	0.046	0.046	0.046	0.046	0.046
		C		0.046	0.046	0.046	0.046	0.046	0.047
Standard error se(δAG)	0.222	A		0.222	0.222	0.222	0.222	0.222	0.221
		B		0.222	0.222	0.222	0.222	0.222	0.221
		C		0.222	0.222	0.222	0.222	0.222	0.221