hdbayes: Bayesian Analysis of Generalized Linear Models with Historical Data (original) (raw)
User-friendly functions for leveraging (multiple) historical data set(s) for generalized linear models. The package contains functions for sampling from the posterior distribution of a generalized linear model using the prior induced by the Bayesian hierarchical model, power prior by Ibrahim and Chen (2000) <doi:10.1214/ss/1009212673>, normalized power prior by Duan et al. (2006) <doi:10.1002/env.752>, normalized asymptotic power prior by Ibrahim et al. (2015) <doi:10.1002/sim.6728>, commensurate prior by Hobbs et al. (2011) <doi:10.1111/j.1541-0420.2011.01564.x>, robust meta-analytic-predictive prior by Schmidli et al. (2014) <doi:10.1111/biom.12242>, the latent exchangeability prior by Alt et al. (2023) <doi:10.48550/arXiv.2303.05223>, and a normal (or half-normal) prior. Functions for computing the marginal log-likelihood under each of the implemented priors are also included. The package compiles all the 'CmdStan' models once during installation using the 'instantiate' package.
Version: | 0.1.1 |
---|---|
Depends: | R (≥ 4.2.0) |
Imports: | instantiate (≥ 0.1.0), callr, fs, formula.tools, stats, posterior, enrichwith, mclust, bridgesampling, mvtnorm |
Suggests: | cmdstanr (≥ 0.6.0), ggplot2, knitr, parallel, rmarkdown, tibble, dplyr |
Published: | 2024-08-22 |
DOI: | 10.32614/CRAN.package.hdbayes |
Author: | Ethan M. Alt [aut, cre, cph], Xinxin Chen [aut], Luiz M. Carvalho [aut], Joseph G. Ibrahim [aut], Xiuya Chang [ctb] |
Maintainer: | Ethan M. Alt |
BugReports: | https://github.com/ethan-alt/hdbayes/issues |
License: | MIT + file |
URL: | https://github.com/ethan-alt/hdbayes |
NeedsCompilation: | yes |
SystemRequirements: | CmdStan (https://mc-stan.org/users/interfaces/cmdstan) |
Additional_repositories: | https://mc-stan.org/r-packages/ |
Materials: | NEWS |
CRAN checks: | hdbayes results |
Documentation:
Downloads:
Linking:
Please use the canonical formhttps://CRAN.R-project.org/package=hdbayesto link to this page.