VBLasso: Bayesian Lasso by Variational Bayes

In lijinsgithub/BLasso: Bayesian Lasso

Description

Fit Bayesian Lasso (Park \& Casella (2008)) by variational Bayes

Usage

VBLasso(x, y, print.it = FALSE)

Arguments

print.it	= TRUE/FALSE (default: FALSE, suppressing to print the number of iterations)
x:	predictor variables (numertic only)
y:	outcome (numertic only)

Value

beta

beta.sig: standard deviation of beta

p.value: p-value for t-test for beta (for variable selection)

sigma2

lambda: penalty, or global scale

convergence: 1/0 converged if convergence = 1

References

Park, Trevor, and George Casella. "The bayesian lasso." Journal of the American Statistical Association 103.482 (2008): 681-686.

Joo, Lijin. "Bayesian Lasso: An Extension for Genome-wide Assoication Study." New York University, 2017

Examples

ex1<-VBLasso(x=data1[,-1], y=data1[,1]); ex1$beta; ex1$lambda; sum(ex1$p.value<0.05); #n of selected variables#

lijinsgithub/BLasso documentation built on May 21, 2019, 6:15 a.m.