Chiang Mai Journal of Science

We consider methods for modeling and comparing incidence rates of adverse events that vary over space, time and demographic characteristics of subjects including gender and age group. We assume that the adverse events and their corresponding population-at-risk denominators are aggregated into a contingency table whose dimensions correspond to the spatial, temporal and demographic factors. We compare regression models for Poisson and negative binomial generalized linear models with zero-corrected log-transformed linear models. These methods are applied to the terrorism events in regions of Southern Thailand that occurred over the period from 2004 to 2010.