On weighted hybrid pairwise likelihood estimation for longitudinal data with LEAR correlation

Pairwise likelihood estimation is widely used to address the intractable full likelihood functions or the difficulty in specifying and verifying the correct high-dimensional joint distribution of data. Weighting has been proposed for increasing efficiency but the choice of weights is not straightforward in most applications. In this paper, we discuss the weight selection for hybrid pairwise likelihood estimation in longitudinal data. Some analytic results are obtained under a multivariate normal distribution with a linear exponent autoregressive (LEAR) correlation structure. We demonstrate the performance of different weighting schemes via simulations and analysis of CD4 data.