annuncio seminario

[Elena Stanghellini <stanghel>]

Il seguente seminario avra' luogo nell'auletta interna del
Dipartimento di Scienze Statistiche dell'Universita' di Perugia,
Via Pascoli,1.

Data: Venerdi' 19 gennaio 2001, ore 15:00.

Speaker: Francesca Chiaromonte, Department of Statistics,
Pennsylvania State University (chiaro@stat.psu.edu).

Titolo: Sufficient Dimension Reduction with Categorical Predictors.

Riassunto:
 
 Following the pioneering work of Ker Chau Li and Dennis Cook, the
 last ten years have witnessed the development of a large
 methodological and applied literature on sufficient dimension 
 reduction for regression.
 
 In problems with a large number of quantitative predictors, 
 sufficient dimension reduction targets a low-dimensional subspace
 (a small number of linear combinations) conveying all the regression 
 information available from the original predictors. Through 
 inference on this subspace, we can visualize, and confine model
 building to, a new regression comprising only a few predictors.
 
 While this approach imposes assumptions much milder than those
 involved in other regression analysis strategies, its scope
 its clearly limited to quantitative predictors. In some recent
 developments, Dennis Cook (Univ. of Minnesota), Bing Li (Pennsylvania
 State Univ.) and myself extended the theoretical core and
 some of the classical inferential machinery of sufficient dimension
 reduction to settings involving both quantitative and categorical
 predictors. Our exercise opens the way to the definition of
 reduction schemes involving constrains, so we refer to it as
 partial sufficient dimension reduction.  
 
 In this talk, I will review classical sufficient dimension reduction
 and present our extension.


**********************************************
Elena Stanghellini 
Dipartimento di Scienze Statistiche
Via A. Pascoli - C.P. 1315 Succ. 1
06100 Perugia (Italy)

Tel +39 075 5855229 or 5855242
Fax +39 075 43242

email: stanghel@stat.unipg.it
home page: http://www.stat.unipg.it/DSS/elena.html
**********************************************

=================================================================

       Forum per la Societa' Italiana di Statistica

                       AVVERTENZA

     il nuovo servizio viene offerto in via sperimentale
     dal Dipartimento di Scienze Statistiche di Perugia
           con il software Majordomo 1.94.5
       istallato su una Workstation UNIX (AIX 4.3.3)

           Ci scusiamo per eventuali disguidi !

=================================================================