[Forum SIS] Annuncio seminario - Geoff McLachlan

[Statlab- UNISA]

************************************************************************

Annuncio Seminario di Statistica

Dipartimento di Scienze Economiche e Statistiche - DiSES

Università degli Studi di Salerno

************************************************************************

 

Martedì 23 settembre 2014 ore 14:30 - Sala del Consiglio DiSES, Stecca 3 -
1° piano

 

Speaker: Geoff McLachlan (University of Queensland, Australia) 

(http://www.maths.uq.edu.au/~gjm/ )

 

Email: g.mclachlan a uq.edu.au 



Titolo: Mixture Models with High-Dimensional Data

 

Abstract: Mixture distributions are being commonly employed these days to
model and cluster heterogeneous data in a wide variety of fields. 

They provide a very flexible approach to clustering data as they allow for
clusters of varying shapes, including asymmetry, and for the presence of
outliers. However, for a large number of variables p, mixtures of normal and
t-component distributions are highly parameterized models since the
component scale matrices consist of

p(p+1)/2 distinct parameters. Moreover, in many problems these days, the
number of observations n is relatively small compared to p. Hence some form
of dimension reduction and/or regularization is needed in such situations.
In this talk we shall discuss some methods for this purpose, focusing on the
use of mixtures of factor models.

 

 

Tutti gli interessati sono cordialmente invitati a partecipare.

 

Laboratorio di Ricerca e Formazione avanzata in Statistica (STATLAB)
<http://www.dises.unisa.it/centri_laboratori/statlab/index> 

Dipartimento di Scienze Economiche e Statistiche - DiSES
<http://www.dises.unisa.it/index>  

 

Università degli Studi di Salerno 

Via Giovanni Paolo II, 132

84084 Fisciano SA

Tel. +39 089 96 3132

 

 



---
Questa e-mail è priva di virus e malware perché è attiva la protezione avast! Antivirus.
http://www.avast.com
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20140915/99a94a76/attachment-0001.html>