[Forum SIS] AVVISO DI SEMINARI dei PROFF. Soren JOHANSEN e Steffen LAURITZEN

[Julia Mortera]

***
                                      *************AVVISO DI  
SEMINARI********************

*mercoledì 3 novembre 2010 aula 19 (2° piano)*

*al Dipartimento**di Economia - Via Silvio D'Amico 77, 00145 Roma*

*si terranno i seguenti seminari:***

***ore 15:00 Søren Johansen - **An analysis of the indicator saturation 
estimator as a robust regression estimator***

*ore 16:15 Steffen Lauritzen - **Graphical Gaussian Models with Symmetries*

***Title: "An analysis of the indicator saturation estimator as a robust 
regression estimator"*

// /Presenter: Søren Johansen, Department of Economics, University of 
Copenhagen and CREATES, University of Aarhus/
**
*Abstract:*
An algorithm suggested by Hendry (1999) for estimation in a regression 
with more regressors than observations, is analyzed with the purpose of 
finding an estimator that
is robust to outliers and structural breaks. This estimator is an 
example of a one-step M-estimator based on Huber.s skip function. The 
asymptotic theory is derived in the
situation where there are no outliers or structural breaks using 
empirical process techniques. Stationary processes, trend stationary 
autoregressions and unit root processes
are considered.The methods are used to analyse the Forward Search 
Algorithm, by Atkinson,Riani, and Ceroli.
//
/_References:_/
1. Johansen, S. and Nielsen, B. (2009) "An analysis of the indicator 
saturation estimator as a robust regression estimator" In N. Shephard 
and J. L. Castle (ed):
The Methodology and Practice of Econometrics: A Festschrift in Honour of 
David F. Hendry, 1.36 Oxford University Press.
2. Johansen, S. and Nielsen, B. (2010) "Discussion of "The Forward 
Search: Theory and Data Analysis", by Atkinson, A.C., Riani, M., and 
Ceroli, A." Journal of the
Korean Statistical Society, 39, 137.145.

*Title: Graphical Gaussian Models with Symmetries
*
/Presenter: Steffen Lauritzen, Head of the Department of Statistics //at 
the University of Oxford **/

Graphical Gaussian Models are typically defined through restricting 
elements of the inverse covariance matrix, aka the concentration matrix, 
to be equal to zero; this corresponds to conditional independence 
restrictions of an undirected graph determined by absence of edges 
between pairs of variables with zero joint concentration. In this way, 
complex multivariate distributions may be described with parsimony. This 
lecture is concerned with further restrictions on the distribution 
determined by symmetry; the symmetry restrictions are obtained by 
partitioning the vertex set into vertex colour classes and the edge set 
into edge colour classes so that parameters corresponding to objects in 
the same colour class are restricted to be identical. The lecture will 
describe examples of such models and discuss their properties. The 
material presented is based on joint work with Søren Højsgaard and 
Helene Neufeld.
_
Reference:_
Højsgaard, S. and Lauritzen, S. L. (2008). Graphical Gaussian models 
with edge and vertex symmetries. /Journal of Royal Statistical Society, 
Series B, /*70, *1005-1027
___________________________________________________________________________________________________________________________________________________________________________

* Sorry for any cross-posting*
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20101029/54c03cf1/attachment.html>