10.5281/zenodo.20724550 (version 0.1.1 archive).Initial CRAN release.
prior_fraction() computes the per-group prior fraction (pooling or
shrinkage factor) for hierarchical models. Accepts a brmsfit object
directly, or a manual path supplying prior precision and likelihood
information for any other fitting engine (Stan, JAGS, etc.).
Supports gaussian, bernoulli, binomial, poisson, and negbinomial families.
Has print and plot methods.
smoothbp_advisor() reports the same Fisher information decomposition for
changepoint random effects in smoothbp fits, and recommends centred vs.
non-centred parameterisation per group. Has print and plot methods.
The underlying geometry (analytic connection, curvature, chain-level coupling
diagnostic, and experimental samplers) is reproduction code for the companion
paper and lives in the source repository's paper/ directory rather than in
the installed package.