An Iterative Weighted Least Squares Procedure for Analysis of Longitudinal Data

We propose an iterative weighted least squares procedure for analysis of longitudinal data. We show that, under very mild conditions, the ability that the procedure converges at an exponential rate tends to one as sample size increases. Furthermore, we show that the limiting estimator is consistent and asymptotically efficient, as expected. The method does not require a parametric covariance structure and/or normality of the data. Finite sample performance of the procedure is studied using a simulated example. A numerical example is considered, in which we use the method to analyze a set of medical data.