STATS_BAYES_SELECTVARS

This extension procedure provides a Bayesian method for selecting the independent variables for linear or generalized linear regression models by comparing their Bayes factors. It uses the ratio of the integrated (i.e., marginal) likelihoods. For linear model coefficients, tt uses an approximation to the Jeffreys-Zellner-Siow prior which uses the Jeffreys prior on sigma and the Zellner-Siow Cauchy prior on the coefficients. For generalized linear coefficients, it uses CCH with alpha = .5. Other priors are available in the underlying BAS package but are not currently provided in this extension.