Extra-Poisson Variation in Log-Linear Models
N. E. Breslow · Journal of the Royal Statistical Society Series C (Applied Statistics) · 1984
A modification of the iterated reweighted least squares scheme of Williams (1982) conveniently accommodates extra‐Poisson variation when fitting log‐linear models to tables of frequencies or rates. The method is applied to the analysis of cancer death rates by age and birth cohort and to testing for mutagenic effects in a standard bioassay.
Cited by 746 works. doi:10.2307/2347661
Authors
References
- Quasi-Likelihood Functions, Generalized Linear Models, and the Gauss-Newton Method
- Extra-Binomial Variation in Logistic Linear Models
- The Estimation of Age, Period and Cohort Effects for Vital Rates
- Regression of Area Mortality Rates on Explanatory Variables: What Weighting is Appropriate?
- Compound Poisson Regression Models
- Statistical analysis of the Ames Salmonella/microsome test.
- Inference Sensitivity for Poisson Mixtures
- TEMPORAL TRENDS IN BREAST CANCER