InvStablePrior: Inverse Stable Prior for Widely-Used Exponential Models

Contains functions that allow Bayesian inference on a parameter of some widely-used exponential models. The functions can generate independent samples from the closed-form posterior distribution using the inverse stable prior. Inverse stable is a non-conjugate prior for a parameter of an exponential subclass of discrete and continuous data distributions (e.g. Poisson, exponential, inverse gamma, double exponential (Laplace), half-normal/half-Gaussian, etc.). The prior class provides flexibility in capturing a wide array of prior beliefs (right-skewed and left-skewed) as modulated by a parameter that is bounded in (0,1). The generated samples can be used to simulate the prior and posterior predictive distributions. More details can be found in Cahoy and Sedransk (2019) <doi:10.1007/s42519-018-0027-2>. The package can also be used as a teaching demo for introductory Bayesian courses.

Version: 0.1.1
Imports: stats, fdrtool, nimble
Published: 2023-08-21
DOI: 10.32614/CRAN.package.InvStablePrior
Author: Dexter Cahoy [aut, cre], Joseph Sedransk [aut]
Maintainer: Dexter Cahoy <dexter.cahoy at>
License: GPL (≥ 3)
NeedsCompilation: no
Materials: README
CRAN checks: InvStablePrior results


Reference manual: InvStablePrior.pdf


Package source: InvStablePrior_0.1.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): InvStablePrior_0.1.1.tgz, r-oldrel (arm64): InvStablePrior_0.1.1.tgz, r-release (x86_64): InvStablePrior_0.1.1.tgz, r-oldrel (x86_64): InvStablePrior_0.1.1.tgz
Old sources: InvStablePrior archive


Please use the canonical form to link to this page.