CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Model selection with missing data Authors:  Sylvain Sardy - University of Geneva (Switzerland) [presenting]
Abstract: Model selection in high-dimensional linear models is considered when some entries of the regression matrix are missing. The goal is to be the least affected by the missing values so as to achieve high true positive rate and low false discovery rate in the search for the true underlying covariates.