Estimation using penalized quasilikelihood and quasi-pseudo-likelihood in Poisson mixed models
We consider two estimation schemes based on penalized quasilikelihood and quasi-pseudo-likelihood in Poisson mixed models. The asymptotic bias in regression coefficients and variance components estimated by penalized quasilikelihood (PQL) is studied for small values of the variance components. We show the PQL estimators of both regression coefficients and variance components in Poisson mixed models have a smaller order of bias compared to those for binomial data. Unbiased estimating equations based on quasi-pseudo-likelihood are proposed and are shown to yield consistent estimators under some regularity conditions. The finite sample performance of these two methods is compared through a simulation study.
KeywordsAsymptotic bias Estimating equations Generalized linear mixed models Laplace expansion Overdispersion Variance components
Unable to display preview. Download preview PDF.
- Hinde J (1982) Compound Poisson regression models. In: Gilchrist Berline R (ed) GLIM 82: Proceedings of the International Conference on Generalised Linear Models. Springer-Verlag, pp 109–121Google Scholar
- Nelder JA and Lee Y (1992). Likelihood, quasi-likelihood and pseudolikelihood: some comparisons. J R Statist Soc B 50: 266–268 Google Scholar
- Schall R (1991). Estimation in generalized linear models with random effects. Biometrika 40: 917–927 Google Scholar