Model Selection in Linear Mixed Effects Models Using SAS PROC MIXED

Although there are disadvantages associated with model building procedures such as backward, forward and stepwise procedures (e.g. multiple testing, arbitrary significance level used in dropping or acquiring variables), many analysts use these procedures and are not aware that alternative modeling selection methods exist. This paper focuses on model selection using the Akaike Information Criteria (AIC) in the case of linear mixedeffects models. AIC’s fundamental concepts are reviewed and two examples are given to demonstrate its use through PROC MIXED. Master-level biostatisticians, epidemiologists, and others who are working with longitudinal data are encouraged to investigate AIC as the tool in modeling repeated measures data.