Clustered binary data with random cluster sizes:a dual poisson modelling approach

摘要

In the analysis of clustered binary data with random cluster sizes, traditional approachesassuming fixed cluster sizes are generally used. Appropriate inference should take account of bothintra-cluster correlation and extra-variation arising from the random cluster sizes. We introduce adual Poisson random effects model for performing appropriate analyses of such data. Our orthodoxbest linear unbiased predictor approach to this model depends only on the first- and second- momentassumptions of unobserved random effects. This approach is illustrated with analyses of seed germina-tion data and developmental toxicity data.