Diagnostic tools for a multivariate negative binomial model for fitting correlated data with overdispersion

We focus on the development of diagnostic tools and an R package called MNB for a multivariate negative binomial (MNB) regression model detecting atypical influential subjects. The is deduced from Poisson mixed in which random intercept follows generalized log-gamma (GLG) distribution. correlated count data leads to that inherits features hierarchical accommodate intraclass correlation occurrence overdispersion simultaneously. asymptotic consistency dispersion parameter estimator depends asymmetry GLG Inferential procedures are simple, although it can provide inconsistent estimates variance when structure misspecified. propose randomized quantile residual checking adequacy derive global local influence measures assess Finally, two applications presented analysis section. code installing used examples exhibited Appendices.