Premium
Bayesian variable selection for logistic regression
Author(s) -
Tian Yiqing,
Bondell Howard D.,
Wilson Alyson
Publication year - 2019
Publication title -
statistical analysis and data mining: the asa data science journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.381
H-Index - 33
eISSN - 1932-1872
pISSN - 1932-1864
DOI - 10.1002/sam.11428
Subject(s) - feature selection , logistic regression , computer science , bayesian probability , bayesian linear regression , selection (genetic algorithm) , variable (mathematics) , artificial intelligence , calibration , statistics , data mining , machine learning , bayesian inference , pattern recognition (psychology) , mathematics , mathematical analysis
A key issue when using Bayesian variable selection for logistic regression is choosing an appropriate prior distribution. This can be particularly difficult for high‐dimensional data where complete separation will naturally occur in the high‐dimensional space. We propose the use of the Normal‐Gamma prior with recommendations on calibration of the hyper‐parameters. We couple this choice with the use of joint credible sets to avoid performing a search over the high‐dimensional model space. The approach is shown to outperform other methods in high‐dimensional settings, especially with highly correlated data. The Bayesian approach allows for a natural specification of the hyper‐parameters.