ba700d58c5f51410VgnVCM100000c2b1d38dRCRDapproved/UMICH/stats/Home/News & Events/Statistics SeminarSeminar Series: Graph estimation with matrix variate normal instances###@###(Fri, 11 Oct 2013)Seminar Series: Graph estimation with matrix variate normal instances###@###(Fri, 11 Oct 2013)411 West Hallstats1381505400000138150540000011:30 AM<p><span style=" color: rgb(34, 34, 34); float: none; line-height: 18.99147605895996px; background-color: rgb(255, 255, 255); font-style: normal; font-weight: normal; font-variant: normal; letter-spacing: normal; word-spacing: 0px; font-size: 13.63636302947998px; text-indent: 0px; display: inline !important; -webkit-text-stroke-width: 0px; font-family: 'Lucida Grande', helvetica, arial, verdana, sans-serif; orphans: auto; text-transform: none; widows: auto; white-space: normal; text-align: start;">Undirected graphs can be used to describe matrix variate distributions. In this talk, I describe new methods for estimating the graphical structures and underlying parameters, namely, the row and column covariance and inverse covariance matrices from the matrix variate data. Under sparsity conditions, we show that one is able to recover the graphs and covariance matrices with a single random matrix from the matrix variate normal distribution. Consistency and the rates of convergence in the operator and the Frobenius norm will be established.</span></p>Njkmcdonbzuniga13813480557978a700d58c5f51410VgnVCM100000c2b1d38d____once11112newnewProfessor Shuheng Zhou, University of Michigan