Bayesian Dynamic Models for Poisson and Binomial Time Series R package DynCount version 0.1.0 Fits Bayesian state-space models for non-Gaussian time series using a latent log-rate (Poisson) or latent logit (binomial) formulation. The latent trajectory follows a first-order random walk or a stationary AR(1) process, sampled by Metropolis-within-Gibbs using the implied Gaussian Markov random field (GMRF) full conditionals. Four innovation structures are supported for the latent increments: constant-variance Gaussian, Student-t, a finite scale mixture of normals, and stochastic volatility. Both families support time-constant zero inflation. The package provides simulation, fitting, forecasting, summary and plotting tools. It implements and extends the methodology of Zens and Bijak (2026) <doi:10.1214/26-AOAS2171>. Software Depends: R (>= 3.5.0) Imports: stats, graphics, grDevices, utils Suggests: stochvol, testthat (>= 3.0.0), knitr, rmarkdown Gregor Zens <zens@iiasa.ac.at> Comprehensive R Archive Network (CRAN) Gregor Zens [aut, cre] MIT + file LICENSE (https://CRAN.R-project.org/package=DynCount/LICENSE) 2026-07-14 application/tgz https://CRAN.R-project.org/package=DynCount doi:10.32614/CRAN.package.DynCount en-GB