This package contains the functions, igg and igg.normalmeans, for implementing Bayesian linear regression, sparse normal means estimation, and variable selection with the inverse gamma-gamma (IGG) prior, as introduced by Bai and Ghosh (2018) <arXiv:1710.04369>
The DESCRIPTION file: Inverse Gamma-Gamma IGG
This package implements the IGG model for sparse Bayesian linear regression and the normal means problem. Our package performs both estimation and model selection. The igg and igg.normalmeans functions also returns the endpoints of the credible intervals (i.e. the 2.5th and 97.5th percentiles) for every single parameter of interest, so that uncertainty quantification can be assessed.
Bai, R. and Ghosh, M. (2018). "The Inverse Gamma-Gamma Prior for Optimal Posterior Contraction and Multiple Hypothesis Testing." Submitted, arXiv:1711.07635.