README.md

In yandorazhang/R2D2: Bayesian Linear Regression Using the R2-D2 Shrinkage Prior

R2D2: Bayesian linear regression using the R2-D2 shrinkage prior

Overview

This package is designed for fitting linear regression model using a Bayesian global and local shrinkage prior, i.e., R2-D2 shrinkage prior. The R2-D2 prior contains two types: conditional R2-D2 and marginal R2-D2. The conditional R2-D2 is a prior conditioned on the the design matrix, while the marginal R2-D2 is free of the design matrix. In general, the marginal R2-D2 shrinkage prior is recommended.

R2D2 Installation

R2D2 software can be installed via Github. To install the latest version of R2D2 package via Github, run following commands in R:

if (!require("devtools"))
  install.packages("devtools")
devtools::install_github("yandorazhang/R2D2",force=T)

Readme

To use the conditional R2-D2 prior, type r2d2cond().

To use the marginal R2-D2 prior, type r2d2marg().

Too use the Dirichlet-Laplace prior proposed in Bhattacharya et al. 2015, type dl().

Citation

Please cite the following paper when you use R2D2:

Zhang, Yan Dora, et al. "Bayesian Regression Using a Prior on the Model Fit: The R2-D2 Shrinkage Prior." Journal of the American Statistical Association (2020): 1-37. (https://doi.org/10.1080/01621459.2020.1825449)

Contact the Author

Author: Yan Dora Zhang, Brian P. Naughton, Howard D. Bondell, Brian J. Reich

Software Developer: Yuxi Cai, Yan Dora Zhang

Software Maintainer: Yan Dora Zhang ([e-mail] or [e-mail])

yandorazhang/R2D2 documentation built on Nov. 23, 2020, 7:05 a.m.