CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Bayesian model selection for semi-parametric models Authors:  Weining Shen - UC Irvine (United States) [presenting]
Abstract: Bayesian analysis is conducted on a class of semi-nonparametric regression models with high-dimensional parametric covariates. In particular, we show (1) strong model selection consistency, where the posterior probability of the true model converges to one; and (2) joint BvM theorem, where the posterior of the selected parametric and nonparametric components jointly converges to a Gaussian vector.