Table 2 Set up of simulation studies
\(\varvec{\alpha }_{\varvec{1}}\) | \(\varvec{\alpha }_{\varvec{2}}\) | \(\varvec{\alpha }_{\varvec{3}}\) | \(\varvec{\alpha }_{\varvec{4}}\) | |
|---|---|---|---|---|
Model 1 (LME with a linear time trend) | ||||
Scenario 1 (low correlation) | 1 | 1 | 0 | 0 |
Scenario 2 (high correlation) | 0 | 0 | 1 | 1 |
Scenario 3 (both groups) | 1 | 1 | 1 | 1 |
Scenario 4 (null case) | 0 | 0 | 0 | 0 |
Scenario 5a (complete null case) | 0 | 0 | 0 | 0 |
Model 2 (LME with a quadratic term for time) | ||||
Scenario 1 (low correlation) | 1 | 1 | 0 | 0 |
Scenario 2 (high correlation) | 0 | 0 | 1 | 1 |
Scenario 3 (both groups) | 0.5 | 0.5 | 0.5 | 0.5 |
Scenario 4 (null case) | 0 | 0 | 0 | 0 |
Scenario 5a (complete null case) | 0 | 0 | 0 | 0 |
Model 3 (LME with a 3-knot spline function for time) | ||||
Scenario 1 (low correlation) | 1 | 1 | 0 | 0 |
Scenario 2 (high correlation) | 0 | 0 | 1 | 1 |
Scenario 3 (both groups) | 1 | 1 | 1 | 1 |
Scenario 4 (null case) | 0 | 0 | 0 | 0 |
Scenario 5a (complete null case) | 0 | 0 | 0 | 0 |