Search Results for author: Yasuyuki Hamura

Found 2 papers, 2 papers with code

Log-Regularly Varying Scale Mixture of Normals for Robust Regression

1 code implementation6 May 2020 Yasuyuki Hamura, Kaoru Irie, Shonosuke Sugasawa

Linear regression with the classical normality assumption for the error distribution may lead to an undesirable posterior inference of regression coefficients due to the potential outliers.

Methodology

On Global-local Shrinkage Priors for Count Data

1 code implementation2 Jul 2019 Yasuyuki Hamura, Kaoru Irie, Shonosuke Sugasawa

Global-local shrinkage prior has been recognized as useful class of priors which can strongly shrink small signals towards prior means while keeping large signals unshrunk.

Methodology

Cannot find the paper you are looking for? You can Submit a new open access paper.