论文信息 - A new package for fitting random effect models.

A new package for fitting random effect models.

Random effects have become a standard concept in statistical modelling over the last decades. They enter a wide range of applications by providing a simple tool to account for such problems as model misspecification, unobserved (latent) variables, unobserved heterogeneity, and the like. One of the most important model classes for the use of random effects is the generalized linear model. Aitkin (1999) noted that “the literature on random effects in generalized linear models is now extensive”, and this is certainly even more true today.

[1] John Hinde,et al. Analyzing Irish suicide rates with mixture models. , 2006 .

[2] J. Hinde,et al. A Note on NPML Estimation for Exponential Family Regression Models with Unspecified Dispersion Parameter , 2006 .

[3] Sophia Rabe-Hesketh,et al. Generalized latent variable models: multilevel, longitudinal, and structural equation models , 2004 .

[4] M. Aitkin. A General Maximum Likelihood Analysis of Variance Components in Generalized Linear Models , 1999, Biometrics.

[5] Murray Aitkin,et al. A general maximum likelihood analysis of overdispersion in generalized linear models , 1996, Stat. Comput..

[6] Bruce G. Lindsay,et al. Computer-assisted analysis of mixtures (C.A.MAN) statistical algorithms , 1992 .

[7] R. Tsutakawa,et al. Estimation of cancer mortality rates: a Bayesian analysis of small frequencies. , 1985, Biometrics.

[8] N. Laird. Nonparametric Maximum Likelihood Estimation of a Mixing Distribution , 1978 .

[9] Murray Aitkin,et al. Fitting overdispersed generalized linear models by non-parametric maximum likelihood. , 1995 .

[10] G. Brush,et al. Components of growth variation in human stature. , 1990, IMA journal of mathematics applied in medicine and biology.