关键词: 变量选择, FIC准则, 混合效应模型

Abstract: Linear mixed-effects (LME) models have been widely used in many scientific fields. This paper studies variable selection for LME models. We derive the focused information criterion (FIC), which selects the model that minimizes the mean squared error (MSE) of the estimator of the focused parameter to improve the efficiency of estimation. The simulation study shows that the proposed FIC performs better than the other commonly used model selection methods.

Key words: Variable Selection, Focused Information Criterion, Mixed-Effects Model