Value for parameter | Reason for determining value |
---|---|
\({\theta }_{0}={\text{log}}\left(0.35/0.65\right)\) | the true log odds ratio of the outcome in the control group (\({W}_{ijk}=0\)) with reference group for binary covariate (\({X}_{ijk}=0\)) for kth individual in ith cluster at jth period. We assume that prevalence of outcome is 15% |
\({\gamma }_{1}=0, {\gamma }_{2}=0,{\gamma }_{3}=0,{\gamma }_{4}=0,{\gamma }_{5}=0\) | No secular trend |
\({\theta }_{1}={\text{log}}\left(1.24\right)\) \({\theta }_{2}={\text{log}}\left(0.33\right)\) \({\theta }_{3}={\text{log}}\left(1.96\right)\) | Assuming minority group is 33%, OTE and main effect of minority status are based on the hypothesis: GoC increases in minority and non-minority patients are from 15 to 30% and from 35 to 40%, respectively |
\(I=8,J=5,m=15\) | Number of clusters, and time steps. Equal sizes for every cluster |
\(ICC=0.1\); \(CAC=0.8\) | For a nested exchangeable correlation structure |