Fast sampling with Gaussian scale-mixture priors in high-dimensional regression. Academic Article

abstract

• We propose an efficient way to sample from a class of structured multivariate Gaussian distributions. The proposed algorithm only requires matrix multiplications and linear system solutions. Its computational complexity grows linearly with the dimension, unlike existing algorithms that rely on Cholesky factorizations with cubic complexity. The algorithm is broadly applicable in settings where Gaussian scale mixture priors are used on high-dimensional parameters. Its effectiveness is illustrated through a high-dimensional regression problem with a horseshoe prior on the regression coefficients. Other potential applications are outlined.

authors

• Bhattacharya, Anirban
• Mallick, Bani

published proceedings

• Biometrika

altmetric score

• 0.5

author list (cited authors)

• Bhattacharya, A., Chakraborty, A., & Mallick, B. K.

citation count

• 59

complete list of authors

• Bhattacharya, Anirban||Chakraborty, Antik||Mallick, Bani K

publication date

• December 2016

publisher

• Oxford University Press (OUP)  Publisher

published in

• Biometrika  Journal

keywords

• Confidence Interval
• Gaussian Scale Mixture
• Global-local Prior
• Shrinkage
• Sparsity

PubMed Central ID

• 28435166

Digital Object Identifier (DOI)

• 10.1093/biomet/asw042

start page

• 985

end page

• 991

volume

• 103

issue

• 4