In statistics, the g-prior is an objective prior for the regression coefficients of a multiple regression. It was introduced by Arnold Zellner. It is a key tool in Bayes and empirical Bayes variable selection.
Assume the are iid normal with zero mean and variance . Let be the matrix with th row equal to . Then the g-prior for is the multivariate normal distribution with prior mean a hyperparameter and covariance matrix proportional to , i.e.,
where g is a positive scalar parameter.
Posterior distribution of 
The posterior distribution of is given as
is the maximum likelihood (least squares) estimator of . The vector of regression coefficients can be estimated by its posterior mean under the g-prior, i.e., as the weighted average of the maximum likelihood estimator and ,
Clearly, as g →∞, the posterior mean converges to the maximum likelihood estimator.
Selection of g
Estimation of g is slightly less straightforward than estimation of . A variety of methods have been proposed, including Bayes and empirical Bayes estimators.
- Zellner, A. (1986). "On Assessing Prior Distributions and Bayesian Regression Analysis with g Prior Distributions". In Goel, P.; Zellner, A. Bayesian Inference and Decision Techniques: Essays in Honor of Bruno de Finetti. Studies in Bayesian Econometrics. 6. New York: Elsevier. pp. 233–243. ISBN 0-444-87712-6.
- George, E.; Foster, D. P. (2000). "Calibration and empirical Bayes variable selection". Biometrika. 87 (4): 731–747. doi:10.1093/biomet/87.4.731.
- Liang, F.; Paulo, R.; Molina, G.; Clyde, M. A.; Berger, J. O. (2008). "Mixtures of g priors for Bayesian variable selection". Journal of the American Statistical Association. 103 (481): 410–423. doi:10.1198/016214507000001337.