250 Current Trends in Bayesian Methodology with Applications
the s ucc e ss of the selection depends on the prior density’s sparsity inducing
properties. In both cases, as one would expect, the c omputation can quickly
become prohibitive. To the best of our knowledge, one of the early papers
that formalized the currently used appro ach to B ayesian selection is [11], who
provide one of the earliest thoroug h details of the use of the g-prior of [26]
in Bayesian Variable Selection. It is important to note however that [12] pro-
vided an even earlier detailed account o n the computational implementation
of Bayesian Variable Selection via the Gibbs Sampling. Arguably, one of the
greatest appeals of the g-prior lies in the fact that it allows the computation ...