Bayesian Variable Selection and Model Averaging using Bayesian
Adaptive Sampling
Description
Package for Bayesian Variable Selection and Model Averaging in linear models and
generalized linear models using stochastic or
deterministic sampling without replacement from posterior
distributions. Prior distributions on coefficients are
from Zellner's g-prior or mixtures of g-priors
corresponding to the Zellner-Siow Cauchy Priors or the
mixture of g-priors from Liang et al (2008)
for linear models or mixtures of g-priors from Li and Clyde
(2019) in generalized linear models.
Other model selection criteria include AIC, BIC and Empirical Bayes estimates of g.
Sampling probabilities may be updated based on the sampled models
using Sampling w/out Replacement or an efficient MCMC algorithm which
samples models using the BAS tree structure as an efficient hash table.
Uniform priors over all models or beta-binomial prior distributions on
model size are allowed, and for large p truncated priors on the model
space may be used to enforce sampling models that are full rank.
The user may force variables to always be included in addition to imposing constraints
that higher order interactions are included only if their parents are
included in the model.
Details behind the sampling algorithm are provided in
Clyde, Ghosh and Littman (2010) .
This material is based upon work supported by the National Science
Foundation under Grant DMS-1106891. Any opinions, findings, and
conclusions or recommendations expressed in this material are those of
the author(s) and do not necessarily reflect the views of the
National Science Foundation.