2 Seminari di Alan Gelfand a Roma

[Luca Tardella <luca.tardella@uniroma1.it>]

Inoltro a tutti gli interessati l'annuncio di 2 seminari
del Professor Alan GELFAND (ISDS, Duke University, Durham, NC, USA)
che si terranno questa settimana a Roma nei giorni di Giovedi' e Venerdi'.



                        avviso di seminario # 1


relatore:                Prof. ALAN E. GELFAND
                  ISDS, Duke University, Durham, NC, USA

titolo:            "Multivariate Spatial Process Modelling"

luogo:                            Sala #34
                                  IV piano
   Dipartimento di statistica, Probabilità e Statistiche Applicate
                          Università "La Sapienza"
                       Piazzale Aldo Moro, 5 - (00185)
                                    ROMA

data e ora:         Giovedi' 6 Ottobre 2005, ore 15:00

abstract:   Models for the analysis of multivariate spatial data are 
receiving
increased attention these days. In many applications it will be 
preferable to work
with multivariate spatial processes to specify such models. A critical 
specification
in providing these models is the cross covariance function. Constructive 
approaches for
developing valid cross-covariance functions offer the most practical 
strategy for
doing this. These approaches include separability, kernel convolution or 
moving average
methods, and convolution of covariance functions. We review these 
approaches but
take as our main focus the computationally manageable class referred to 
as the linear model
of coregionalization (LMC). We introduce a fully Bayesian development of 
the LMC.
We offer clarification of the connection between
joint and conditional approaches to fitting such models including prior 
specifications.
However, to substantially enhance the usefulness of such modelling we 
propose the notion
of a spatially varying LMC (SVLMC) providing a very rich class of 
multivariate nonstationary
processes with simple interpretation.
We illustrate the use of our proposed SVLMC with application to more 
than 600 commercial
property transactions in three quite different real estate markets, 
Chicago, Dallas and
San Diego. Bivariate nonstationary process models are developed for 
income from and selling
price of the property.


          tutti gli interessati sono invitati a partecipare






                        avviso di seminario # 2



relatore:                Prof. ALAN E. GELFAND
                  ISDS, Duke University, Durham, NC, USA

titolo:            "Explaining Species Level Diversity
                Through Species Level Hierarchical Modeling"

luogo:                            Sala #34
                                  IV piano
   Dipartimento di statistica, Probabilità e Statistiche Applicate
                          Università "La Sapienza"
                       Piazzale Aldo Moro, 5 - (00185)
                                    ROMA

data e ora:         Venerdi' 7 Ottobre 2005, ore 12:00


abstract:   Understanding spatial patterns of species diversity and the 
distributions
of individual species is a consuming problem in biogeography and 
conservation.
A dataset from the Cape Floristic Region in South Africa provides some 
60,000 records
to model biodiversity patterns. Model development is focussed on a one 
minute grid-cell
scale yielding roughly 37,000 cells for the region. Using a Bayesian 
framework, we
develop a two-stage spatially explicit hierarchical logistic regression. 
Stage one
models the potential presence or absence of a species in a grid cell 
while stage two
models the probability of observing each species in each cell given that 
it is present.
Sampling intensity varies widely over the region. In addition, portions 
of the landscape
have been transformed due to human intervention. The model provides 
novel measures of
biodiversity. These measures capture notions such as species range, 
species prevalence,
species richness, species abundance, species diversity, and species 
turnover. We discuss
the resultant analysis and provide illustrative maps to display 
estimated biogeographic
patterns across the region. The work is ongoing but our initial findings 
show that the
model performs remarkably well.



-- 
Luca Tardella
Dipartimento di Statistica, Probabilità
e Statistiche Applicate
Università "La Sapienza" - ROMA - 00185
email: luca.tardella@uniroma1.it
tel: 06-4991-0370
fax: 06-4959241


=============================================================================
NOTA:
Le norme per utilizzare il forum SIS e le istruzioni per iscrizione
e cancellazione sono disponibili all'indirizzo
.
.   http://w3.uniroma1.it/sis/forum.asp
.
L'archivio di tutti i messaggi (aggiornato al giorno precedente)
e' disponibile all'indirizzo
.
.   http://www.stat.unipg.it/cgi-bin/wilma/sis/
.
=============================================================================