mastatlab / lips

Scalable Bayesian model averaging through local information propagation

Scalable Bayesian model averaging through local information propagation

This package the LIPS algorithm for scalable Bayesian model averaging on linear regression models. This package uses the C++ library smctc for implementing the sequantial Monte Carlo samplers.

Install

The package can be installed on Linux and Mac using devtools:

install.packages('devtools')
library('devtools')
devtools::install_github('MaStatLab/LIPS')

Use

There are three functions in this package, and their descriptions are provided in the help files

ans = lips(Y,X)
ans = lips_hd(Y,X)
ans = lips_hd_pred(Y,X)

Reference

Ma L. (2015) Scalable Bayesian model averaging through local information propagation. Journal of the American Statistical Association, Theory and Methods. Vol. 110, No. 510, 795-809.