Evaluation Of The Pairwise Approach For Fitting Joint Linear Mixed Models

Dollar longitudinal data (tracking data) is a group were tested in a variable measured over time. Longitudinal data analysis has sophisticated statistical models and statistical inference methods. Multivariate longitudinal data is a group were tested in a number of variables, a number of time measurements. In order to study repeated measurement data of children's number of behavioral indicators,this paper proposes reasonable model and constructions simulation data, uses pairwise method, evaluates this method.Which is from Fieuws. S.&; G. Verbeke (2006) Joint modeling of the proposed pair inspired way of thinking. This paper uses pairwise joint modeling approach, simulates 1000 times and obtains fixed effects estimates of the parameters. In the framework of large sample theory, comparing joint, modeling with the full model, parameters of efficiency loss can be very small. Even in the different responses of the variables have the same case of fixed effectsparameters, parameter estimates and the true value is still very close.