The analysis of multivariate longitudinal data using multivariate marginal models

Longitudinal studies often involve multiple outcomes measured repeatedly from the same subject. The analysis of multivariate longitudinal data can be challenging due to its complex correlated nature. In this paper, we develop multivariate marginal models in longitudinal studies with multiple response variables, and improve parameter estimation by incorporating informative correlation structures. In theory, we show that the proposed method yields a consistent and efficient estimator which follows an asymptotic normal distribution. Monte Carlo studies indicate that the proposed method performs well in the sense of reducing bias and improving estimation efficiency. In addition, the proposed approach is applied to a real longitudinal data example of transportation safety with different response families.