Linear and Non-Linear Mixed Models in Longitudinal Studies andComplex Survey Data

[摘要] Correlated data are fairly common in health and social sciences. Forexample, clustered data arise when subjects are nested in clusters suchas classrooms, hospitals, and neighborhoods; while longitudinal dataresult from multiple measures for the same subject over long period oftime; whereas repeated measures data are involved in multiplemeasurements for the same subject over time or other dimension.Observations for the same cluster/subject are likely to be correlated(non-independent). Mixed models (also known as multilevel modelsor hierarchical models) including both fixed eوٴects and random eوٴectshave been developed to deal with correlated data [1-3]. Нe randomeوٴect of one variable has a prior distribution with variance and variesrandomly within the population; whereas the fixed eوٴect of thevariable is the average eوٴect in the entire population, expressed by theregression coeٹcient.