Joint Modeling for Zero-Inflated Beta-Binomial and Normal Responses

We present a new joint model with random effects for the correlated count with extra zero and continuous responses. In this model, we assume a Zero-Inflated Beta-Binomial distribution for the analysis of overdispersed binomial variable and a normal distribution for the analysis of continuous response. Furthermore, a full model likelihood function approach is used to obtain maximum likelihood estimates of the model parameters. We also evaluate the proposed model using the Monte Carlo simulation method. Finally, we fit the model to real data to find effective factors on mixed responses.