Software

Research software developed and maintained by Theophilus Anim Bediako.

BUCM

Bayesian Underreported Count Modeling

R package · Version 1.0.0 · Author and maintainer

I am the author and maintainer of BUCM, an R package for Bayesian analysis of one- and two-sample count data subject to underreporting. It supports Poisson, negative binomial, and zero-inflated Poisson models using validation data or informative priors.

BUCM provides JAGS-based MCMC estimation, comparisons with models that assume complete reporting, posterior summaries, diagnostic plots, and model comparison using DIC, WAIC, and PSIS-LOO. The documentation includes worked examples and an adverse-event application.

Documentation · User guide · Function reference · Source — v1.0.0

Corresponding manuscript

Anim Bediako, T., Roberman, J. L., Barth, J., & Stamey, J. D. BUCM: An R package for Bayesian modeling of one- and two-sample count data with underreporting. Submitted for publication

BUCM implements the methods described in this manuscript.