The New Approach to Guide the Selection of the Covariance Structure in Mixed Model

This article considers the analysis of experiment of repeated measures design that is used frequently in medical research. We propose new approach could be used to guide the selection of the covariance structure. We used simulation study to compare six model selection criteria in terms of their ability to identify the right correlated error model with the help of the new approach. The comparison of the six model selection criteria were in terms of their percentage of number of times that they identify the right covariance structure with the help of the new approach. The simulation results indicate that overall, the new proposed approach showed very good performance with all the six model selection criteria where the AICC and BIC criteria provided the best performance for all the study cases. [17] [25] The main result of our article is that we recommend using the new proposed approach with AICC or BIC criteria as a standard procedure to identify the right correlated error model.