Effect
Parameter
MCAR
MAR
MNAR
Measurement model
drug1 intercept
β
01
13.91 (0.92)
13.91 (0.92)
13.90 (0.92)
drug2 intercept
β
02
-3.67 (1.30)
-3.67 (1.30)
-3.71 (1.30)
drug3 intercept
β
03
0.69 (1.30)
0.69 (1.30)
0.61 (1.32)
drug1 slope
β
11
1.20 (0.17)
1.20 (0.17)
1.24 (0.17)
drug2 slope
β
12
1.51 (0.19)
1.51 (0.19)
1.60 (0.20)
drug3 slope
β
13
1.35 (0.18)
1.35 (0.18)
1.38 (0.18)
Variance model
Random-intercept variance
d
8.99 (2.63)
8.99 (2.63)
8.99 (2.63)
Serial process variance
τ
2
3.41 (0.56)
3.41 (0.56)
3.35 (0.57)
Serial process correlation
ρ
1.00 (0.00)
1.00 (0.00)
1.00 (0.01)
Measurement error variance
σ
2
0.74 (0.12)
0.74 (0.12)
0.76 (0.12)
-2
l
596.99
591.43
595.56
Table 4:
Multi-centre data: Maximum likelihood for the parameter estimates (standard errors) under MCAR, MAR, and MNAR assumptions without covariate in the dropout model.
