[Forum SIS] DEC Statistics Seminar (Bocconi)- October 22nd

[Isadora Antoniano]

Dear all,
We are glad to announce the next DEC statistics seminar.
Regards,
Raffaella Piccarreta and Isadora Antoniano


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

The DEC seminar schedule is available at
http://www.unibocconi.eu/statseminar



*Thursday, October 22nd*

Mike West
(Duke University)



*"Structured Dynamic Graphical Models & Scaling Multivariate Time Series
Methodology"*

Abstract

This talk links to some of the history and recent developments in time
series using concepts of graphical modelling, emphasizing  scaling analyses
to time series with increasing numbers of variables, flexible multivariate
volatility models, and model evaluation with respect to uncertainty about
graphical model structures.  This includes generalizations of
multi-regression dynamic linear models, referred to as  (a) dynamic
dependency network models, which involve underlying, directed acyclical
graphical model structures, and (b) the more general class of simultaneous
graphical dynamic models, based on underlying undirected graphs.  Some
aspects of model specification, fitting and computation will be noted,
including the use of importance sampling and variational Bayes methods to
implement sequential analysis and forecasting, and exploitation of
GPU-based parallel computing for implementations with higher-dimensional
series.
Applied contexts and illustrations come from financial time series
analysis, multi-step forecasting, and associated Bayesian decision analysis
in portfolio studies.
This work draws on current and research with Lutz Gruber and Zoey Zhao.
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20151016/31a5ca28/attachment-0001.html>
-------------- parte successiva --------------
Un allegato non testuale è stato rimosso....
Nome:        M. West 22.10.15.pdf
Tipo:        application/pdf
Dimensione:  44372 bytes
Descrizione: non disponibile
URL:         <http://www.stat.unipg.it/pipermail/sis/attachments/20151016/31a5ca28/attachment-0001.pdf>