Title: Simultaneous variable and covariance selection with the multivariate spike-and-slab lasso
Authors: Sameer Deshpande - Massachusetts Institute of Technology (United States) [presenting]
Veronika Rockova - University of Chicago (United States)
Edward George - University of Pennsylvania (United States)
Abstract: A Bayesian procedure is proposed for simultaneous variable and covariance selection using continuous spike-and-slab priors in multivariate linear regression models where $q$ possibly correlated responses are regressed onto $p$ predictors. Rather than relying on a stochastic search through the high-dimensional model space, we develop an ECM algorithm similar to the EMVS procedure targeting modal estimates of the matrix of regression coefficients and residual precision matrix. Varying the scale of the continuous spike densities facilitates dynamic posterior exploration and allows us to filter out negligible regression coefficients and partial covariances gradually. Our method is seen to substantially outperform regularization competitors on simulated data.