A comparative study of fixed effects models and random intercept/slope models as a special case of linear mixed models for repeated measurements

Abstract

Any dataset in which subjects are measured repeatedly over time or space can be described as repeated measurements data. A linear mixed model (LMM) is a powerful method for analyzing repeated measurements data. It is made up of two components. The first component consists of a regression model for the average response over time and the effects of covariates on this average response. The second component provides a model for the pattern of covariances or correlations between the repeated measurements. In this study, a comparative evaluation of fixed effects models with random intercept models and random intercept and slope models as a special case of random effects models from linear mixed models are taken into consideration and the superiority of random intercept and slope models allow to modeling possible heterogeneity in intercepts and in slopes of the individual's own regression line for repeated measurements data is emphasized.