Package: ZINB.GP 1.0.0

Hsin-Hsiung Huang

ZINB.GP: Bayesian Zero-Inflated Negative Binomial Gaussian Process Models

Fits Bayesian zero-inflated negative binomial regression models with Gaussian process random effects for spatial, temporal, or spatiotemporal count data. Provides Markov chain Monte Carlo sampling, configurable random effects in the zero-inflation and count components, and posterior predictive draws. Implements a full GP version of the methods described by He and Huang (2024) <doi:10.1016/j.jspi.2023.106098>.

Authors:Mahlon Scott [aut], Qing He [aut], Hsin-Hsiung Huang [aut, cre, cph]

ZINB.GP_1.0.0.tar.gz
ZINB.GP_1.0.0.zip(r-4.7-any)ZINB.GP_1.0.0.zip(r-4.6-any)ZINB.GP_1.0.0.zip(r-4.5-any)
ZINB.GP_1.0.0.tgz(r-4.6-any)ZINB.GP_1.0.0.tgz(r-4.5-any)
ZINB.GP_1.0.0.tar.gz(r-4.7-any)ZINB.GP_1.0.0.tar.gz(r-4.6-any)
ZINB.GP_1.0.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
ZINB.GP/json (API)

# Install 'ZINB.GP' in R:
install.packages('ZINB.GP', repos = c('https://kingjms1.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/kingjms1/gp_zinb_r/issues

Pkgdown/docs site:https://kingjms1.github.io

On CRAN:

Conda:

5.15 score 2 stars 488 downloads 5 exports 20 dependencies

Last updated from:fb7ca2f852. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK183
source / vignettesOK177
linux-releaseOK123
macos-releaseOK202
macos-oldrelOK203
windows-develOK142
windows-releaseOK157
windows-oldrelOK133
wasm-releaseOK147

Exports:kernelmake_prediction_inputsmake_y_Vs_Vtnoise_mixZINB_GP

Dependencies:BayesLogitcliexpmgenericsglueLaplacesDemonlatticelifecyclemagrittrMASSMatrixmsmmvtnormpillarpkgconfigrlangsurvivaltibbleutf8vctrs

Simulating and Fitting a Spatiotemporal ZINB-GP Model
The model | Build a correctly paired space-time design | Baselines and intercepts | Draw latent effects and counts | Fit the full model | Predict at new locations and times | Interpret posterior draws
Rendered fromsimulating-zinb-gp.Rmdusingknitr::rmarkdown

Last update: 2026-07-30
Started: 2026-07-29

Oregon Landslides: A Spatiotemporal ZINB-GP Case Study
Scope of the analysis | Read and align the panel | Describe the data before fitting | Distance matrices and model fit | Fixed effects | Reconstruct and recenter baseline effects | Posterior fitted means | MCMC wrappers and effective sample size
Rendered fromoregon-landslides.Rmdusingknitr::rmarkdown

Last update: 2026-07-29
Started: 2026-07-29

Readme and manuals

Help Manual

Help pageTopics
gp_param_boundsgp_param_bounds
Squared Exponential Kernelkernel
Build Inputs for Prediction at New GP Coordinatesmake_prediction_inputs
make_y_Vs_Vtmake_y_Vs_Vt
mvn_sample_svdmvn_sample_svd
Mix a Matrix with Identity Noisenoise_mix
nullchecknullcheck
Predict at New Locations and Timespredict.zinb_gp_fit
Numerically stable sigmoid functionsigmoid
solve_svdsolve_svd
update_ls_sigma_noiseupdate_ls_sigma_noise
ZINB_GPZINB_GP
ZINB_GP_countZINB_GP_count
ZINB_GP_inflationZINB_GP_inflation
ZINB_GP_origZINB_GP_orig
ZINB_GP_spatialZINB_GP_spatial
ZINB_GP_spatial_countZINB_GP_spatial_count
ZINB_GP_spatial_inflationZINB_GP_spatial_inflation