Fully Bayes factors with a generalized g-prior
Authors:
- Yuzo MaruyamaYMYuzo Maruyama
- Edward I. GeorgeEGEdward I. George
Preprint
Abstract
For the normal linear model variable selection problem, we propose selection criteria based on a fully Bayes formulation with a generalization of Zellner's $g$-prior which allows for $p>n$. A special case of the prior formulation is seen to yield tractable closed forms for marginal densities and Bayes factors which reveal new model evaluation characteristics of potential interest.
Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Empirical Bayes vs. fully Bayes variable selectionJournal of Statistical Planning and Inference, 2008
- Mixtures of g Priors for Bayesian Variable SelectionJournal of the American Statistical Association, 2008
- The Adaptive Lasso and Its Oracle PropertiesJournal of the American Statistical Association, 2006
- A new class of generalized Bayes minimax ridge regression estimatorsThe Annals of Statistics, 2005
- Benchmark priors for Bayesian model averagingJournal of Econometrics, 2001
- Calibration and empirical Bayes variable selectionBiometrika, 2000
- The Risk Inflation Criterion for Multiple RegressionThe Annals of Statistics, 1994
- Minimax Ridge Regression EstimationThe Annals of Statistics, 1980
- Proper Bayes Minimax Estimators of the Multivariate Normal MeanThe Annals of Mathematical Statistics, 1971