No Subject

[Fulvia Mecatti <fulvia.mecatti@unimib.it>]

Per visualizzare questo avviso via web:

http://www.statistica.unimib.it/utenti/mecatti/varie/web_ad.htm

______________________________________________________________

DIPARTIMENTO di STATISTICA,  UNIVERSITA' di MILANO-BICOCCA
DOTTORATO di RICERCA in STATISTICA

AVVISO di SHORT COURSE
_______________________________________________________________

Multivariate Analysis with Fewer Observations  Than The Dimension
Prof. Muni S. Srivastava
Università di Toronto - Canada

24 Maggio 2055 h. 10:00-13:00 Aula Tesi Fac.di Sc.Statistiche (U7, II
piano)
25 Maggio 2005 h. 10:00-13:00 Aula 9 (U7, piano terra)
26 Maggio 2005 h. 10:00-13:00 Aula Tesi Fac.di Sc.Statistiche (U7, II
piano)

Abstract
Recent advances in technology to obtain DNA microarrays have made
it possible to measure quantitatively the expression of thousands of
genes. These 
observations are, however, correlated to each other as the genes are from
the same 
subject.Since the number of subjects available for taking the
observations are few 
and much smaller than the number of genes, multivariate theory needs to
be developed
for this new situation. In this series of talks, I describe these new
theories for 
testing the equality of mean vectors, verifying the assumptions on the
covariance 
matrix, classification and discrimination, and prediction. The theory is
built on 
the existing methodology, and thus they will be discussed first in the
beginning 
of each talk. Analyses of some microarray datasets are given. Topics will
include:
1.Tests for mean vectors including MANOVA
2.Tests on Covariance Matrix
3.Classification and Discrimination
4.Prediction in regression model

Per informazioni e conferma partecipazione:
fulvia.mecatti@unimib.it

Come raggiungere Milano-Bicocca:

http://www.unimib.it/ateneo/orientarsi/

Fulvia Mecatti
Dipartimento di Statistica
02.6448.5872