Premium
Two ways of modelling overdispersion in non‐normal data
Author(s) -
Lee Y.,
Nelder J. A.
Publication year - 2000
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/1467-9876.00214
Subject(s) - overdispersion , quasi likelihood , poisson distribution , statistics , variance (accounting) , count data , econometrics , dispersion (optics) , normal distribution , mathematics , a priori and a posteriori , maximum likelihood , physics , economics , philosophy , accounting , epistemology , optics
For non‐normal data assumed to have distributions, such as the Poisson distribution, which have an a priori dispersion parameter, there are two ways of modelling overdispersion: by a quasi‐likelihood approach or with a random‐effect model. The two approaches yield different variance functions for the response, which may be distinguishable if adequate data are available. The epilepsy data of Thall and Vail and the fabric data of Bissell are used to exemplify the ideas.