Avviso di Seminario - Universita' di Pavia

[Leonardo Bottolo <lbottolo@eco.unipv.it>]

UNIVERSITA' DEGLI STUDI DI PAVIA
DIPARTIMENTO DI ECONOMIA POLITICA E METODI QUANTITATIVI
DIPARTIMENTO DI MATEMATICA "F.CASORATI"
DOTTORATO DI RICERCA IN MATEMATICA E STATISTICA
***********************************************************************


                         AVVISO DI SEMINARIO


Martedi' 13 maggio, alle ore 13.00, nell'aula del Consiglio di Facolta',
Dipartimento di Economia Politica e Metodi Quantitativi (via San Felice 5,
Pavia)


                          PETROS DELLAPORTAS
                      (Department of Statistics
             Athens University of Economics and Business)
             

terra' un seminario dal titolo:


  LARGE CONTEMPORANEOUS ARMA STRUCTURES FOR TIME-VARYING COVARIANCES
              WITH APPLICATIONS TO QUANTITATIVE FINANCE



Abstract:

We use ARMA structures to model contemporaneous correlations amongst 
very many time series. The ensuing method is related to and compares
favourably with the (dynamic) factor models, multivariate GARCH and
stochastic volatility models developed in the last decade in the area
of quantitative finance. As in factor models, the method has the 
drawback of producing results that are unique up to an orthogonal 
rotation (permutation of the variables in a vector of responses).
However, it has the added advantage that the parameters are meaningful
and interpretable as the coefficients of the observables and their
innovations rather than as the factor loadings of some latent factors.
The key theoretical tool used is the modified Cholesky decomposition
of a covariance matrix rather than its traditional spectral decomposition.
Regressogram of the components of the modified Cholesky decomposition is
used, in a manner similar to the correlogram in time series, to identify
parsimonious models for the large number of parameters employing 
parametric, non-parametric and Bayesian methods. This is a joint work with
Mohsen Pourahmadi, Division of Statistics, Northern Illinois University,
USA.





Durante il seminario sarà offerto ai partecipanti un piccolo rinfresco.


***********************************************************************
Per ulteriori informazioni riguardo ai Seminari di Probabilita' e
Statistica e' possibile scrivere a:

lbottolo@eco.unipv.it

oppure consultare la pagina WEB dei seminari:

http://economia.unipv.it/eco-pol/semstat/semstat.htm

-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/