Predictions of machine learning with mixed-effects in analyzing longitudinal data under model misspecification

Abstract We consider predictions in longitudinal studies, and investigate the well known statistical mixed-effects model, piecewise linear model six different popular machine learning approaches: decision trees, bagging, random forest, boosting, support-vector neural network. In order to correlated data learning, effects is combined into traditional tree methods forest. Our focus performance of modelling especially cases misspecification fixed effects. Extensive simulation studies have been carried out evaluate using a number criteria. Two real datasets from are analysed demonstrate our findings. The R code dataset freely available at https://github.com/shuwen92/MEML .