[Forum SIS] Seminario Prof. Consonni 21.10.2011

[Dip. Scienze Statistiche - Mi]

Il Prof. B. Vittorio Frosini, Direttore del Dipartimento di Scienze statistiche dell'Università Cattolica del Sacro Cuore di Milano, ha il piacere di informarLa che il Prof. Guido Consonni, dell'Università degli studi di Pavia, terrà un seminario dal titolo:



Balanced Objective Bayesian Comparison

of Nested Models for Discrete Data


Venerdì 21 Ottobre dalle ore 11,30

in Aula G 052

Università Cattolica del Sacro Cuore - Milano

Largo Gemelli,1 Edificio Lanzone 18



Segue Abstract



Si prega di confermare la propria presenza inviando una e-mail a: dip.scienzestatistiche at unicatt.it<mailto:dip.scienzestatistiche at unicatt.it>
Il convegno è gratuito, ma l'iscrizione necessaria ai fini organizzativi.
Per ulteriori informazioni rivolgersi alla Segreteria Organizzativa:
Sig.ra Barbara Villa
Tel. 02-7234.2647 - Fax 02-7234.3064
e-mail: dip.scienzestatistiche at unicatt.it




Abstract
When two nested models are compared, using a Bayes factor, from an objective standpoint, two seemingly conflicting issues emerge at the time of choosing parameter priors under the two models. On the one hand, for moderate sample sizes, the evidence in favor of the smaller model can be inflated by diffuseness of the prior under the larger model.
On the other hand, asymptotically, the evidence in favor of the smaller model typically accumulates at a slow rate. With reference to discrete data models, we show that these two issues can be dealt with jointly, by combining intrinsic priors and non-local priors in a new unified class of priors. We illustrate our ideas in a running Bernoulli example and then apply them to test the equality of two proportions.

This is joint work with Jonathan J. Forster (University of Southampton, UK) and Luca La Rocca (University of Modena and Reggio Emilia).



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20111011/c93c89bf/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 2950 bytes
Desc: image001.jpg
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20111011/c93c89bf/attachment-0001.jpg>