[Forum SIS] Seminar

[Simone Padoan]

Dear All,
We are glad to announce the following talk

*************************************************
Thursday, November 7th, 2013, 
DEC - Statistics Seminar - Università Bocconi,
Room 3-E4-SR03, Via Rontgen 1 - 3rd floor
Time: 12:30pm

(The complete DEC seminars schedule is available at             
                  http://www.unibocconi.eu/statseminar)
*************************************************


Alfred Hero (University of Michigan)


Kronecker PCA for spatio-temporal covariance analysis


 Abstract 
Similar to standard principal component analysis (PCA), Kronecker PCA approximates a high dimensional sample covariance matrix as a small sum of matrices of simple form, called the principal components. However, unlike standard PCA, in Kronecker PCA each principal component is a matrix that can be expressed as a Kronecker product of two lower dimensional matrices. The number of Kronecker components of the decomposition, called the separation rank, plays a similar role as the algebraic rank in PCA. The components are obtained by computing the singular value decomposition (SVD) of a rectangular matrix that is derived from the sample covariance matrix. The talk will start by discussing Kronecker product estimation in the matrix variate normal model and in sparse versions of this model. It will then introduce Kronecker PCA along with associated high dimensional convergence rates. It will conclude with an illustration on spatio-temporal analysis of meteorological data and future perspectives.

-----------------------------------------------

Dr. Simone Padoan, PhD.
Assistant Professor

Department of Decision Sciences
Bocconi University of Milan
via Roentgen, 1 20136 Milano - Italy
tel. +39 02 5836 5368
Email: simone.padoan a unibocconi.it
Webpage: http://faculty.unibocconi.it/simonepadoan
-----------------------------------------------

-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20131106/11e48256/attachment.html>