BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20141107T105119Z
DTSTART:20141112T160000Z
DTEND:20141112T173000Z
SUMMARY:Mitchell Centre Seminar Series: Alberto Caimo
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}bbf-i1asbx9
c-gtthg6
DESCRIPTION:HBS G6\n\nAlberto Caimo\n\nEfficient Bayesian computation for
exponential random graph models\nPowerful ideas recently appeared in th
e literature are adjusted and combined to design improved samplers for d
oubly intractable target distributions with a focus on Bayesian exponent
ial random graph models. Different forms of adaptive Metropolis-Hastings
proposals (vertical\, horizontal and rectangular) are tested and merged
with the delayed rejection (DR) strategy with the aim of reducing the v
ariance of the resulting MCMC estimators for a given computational time.
The DR is modified in order to integrate it within the approximate exch
ange algorithm (AEA) to avoid the computation of intractable normalising
constant that appears in exponential random graph models. This gives ri
se to the AEA+DR: a new methodology to sample doubly intractable distrib
utions that dominates the AEA in the Peskun ordering leading to MCMC est
imators with a smaller asymptotic variance.\n
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:G6\, Humanities Bridgeford Street\, Manchester
END:VEVENT
END:VCALENDAR