powerprior: Conjugate Power Priors for Bayesian Analysis of Normal Data

Implements conjugate power priors for efficient Bayesian analysis of normal data. Power priors allow principled incorporation of historical information while controlling the degree of borrowing through a discounting parameter (Ibrahim and Chen (2000) <doi:10.1214/ss/1009212519>). This package provides closed-form conjugate representations for both univariate and multivariate normal data using Normal-Inverse-Chi-squared and Normal-Inverse-Wishart distributions, eliminating the need for MCMC sampling. The conjugate framework builds upon standard Bayesian methods described in Gelman et al. (2013, ISBN:978-1439840955).

Version:	1.0.0
Imports:	stats, MASS, LaplacesDemon, ggplot2, shiny, shinydashboard, shinyjs, DT, dplyr, tidyr, rlang
Suggests:	testthat (≥ 3.0.0), rmarkdown
Published:	2025-11-11
DOI:	10.32614/CRAN.package.powerprior (may not be active yet)
Author:	Yusuke Yamaguchi [aut, cre], Yifei Huang [aut]
Maintainer:	Yusuke Yamaguchi <yamagubed at gmail.com>
License:	MIT + file LICENSE
NeedsCompilation:	no
Materials:	README, NEWS
CRAN checks:	powerprior results

Documentation:

Reference manual:

powerprior.html , powerprior.pdf

Downloads:

Package source:	powerprior_1.0.0.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: powerprior_1.0.0.zip
macOS binaries:	r-release (arm64): powerprior_1.0.0.tgz, r-oldrel (arm64): powerprior_1.0.0.tgz, r-release (x86_64): powerprior_1.0.0.tgz, r-oldrel (x86_64): powerprior_1.0.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=powerprior to link to this page.