B1611
Title: Sparse matrix estimation based on greedy algorithms and information criteria
Authors: Hsueh-Han Huang - Academia Sinica (Taiwan) [presenting]
Abstract: The problem of estimating the covariance matrix of serially correlated vectors is considered, whose dimension is allowed to be much larger than the sample size. Using the orthogonal greedy algorithm (OGA) together with a high-dimensional Akaikes information criterion (HDAIC), it is proposed to estimate the matrix and to show that the proposed estimate is rate optimal under a sparsity condition more flexible than those in the existing literature. When the covariance matrix is bandable, a banding/tapering estimate is introduced whose parameters are chosen by a novel information criterion. The rate optimality of the latter estimate is also established.
