Zero-Dependent Bivariate Poisson Distribution with Applications

The bivariate Poisson model is the most widely used for counts, and in recent years, several regression models have been developed order to analyse two response variables that are possibly correlated. In this paper, a particular class of model, from Bernoulli will be presented investigated. proposed use dependence parameters can positively negatively correlated data, whereas more well-known models, such as Holgate’s only data. As result, contributes improving properties common models. Furthermore, some new outlined. method maximum likelihood moment were estimate model. Additionally, real data healthcare utilization sector used. case utilization, between may positive or negative assess performance comparison traditional count All computations graphs shown paper produced using R programming language.