摘要：We propose a versatile class of multiplicative generalized linear longitudinal mixed models (GLLMM) with additive dispersion components, based on explicit modelling of the covariance structure. The class incorp orates a longitudinal structure into the random e.ects models and retains a marginal as well as a conditional interpretation. The estimation procedure is based on a computationally e.cient quasi-score method for the regression parameters combined with a REML-like bias-corrected Pearson estimating function for the dispersion and correlation parameters. This avoids the multidimensional integral of the conventional GLMM likelihood and allows an extension of the robust empirical sandwich estimator for use with both association and regression parameters. The metho d is applied to a set of otholit data, used for age determination of fish.
关键词：Bias correction ; Comp onents of dispersion ; Generalized estimating ; equation ; Pearson estimating function ; REML ; Multiplicative mixed mo dels ; ; Tweedie distribution
