BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Bayesian modelling of Dupuytren disease using Gaus
sian copula graphical models - Reza Mohammadi (Uni
versiteit van Tilburg\; University of Groningen)
DTSTART;TZID=Europe/London:20160826T094000
DTEND;TZID=Europe/London:20160826T102000
UID:TALK67067AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/67067
DESCRIPTION:Co-authors: Fentaw Abegaz (University of Lie
ge\, Belgium)\, Edwin van den Heuvel (Eindhoven U
niversity of Technology\, The Netherlands)\, Erns
t Wit (University of Groningen\, The Netherlands)
Dupuytren disease is a fibr
oproliferative disorder with unknown etiology that
often progresses and eventually can cause perman
ent contractures of the affected fingers. In this
talk\, we provide a computationally efficient Bay
esian framework to discover potential risk factor
s and investigate which fingers are jointly affec
ted. Our Bayesian approach is based on Gaussian co
pula graphical models\, which provide a way to di
scover the underlying conditional independence st
ructure of variables in multivariate mixed data. I
n particular\, we combine the semiparametric Gaus
sian copula with extended rank likelihood to analy
se multivariate mixed data with arbitrary margina
l distributions. For the structural learning\, we
construct a computationally efficient search algo
rithm using a trans-dimensional MCMC algorithm ba
sed on a birth-death process. In addition\, to ma
ke our statistical method easily accessible to oth
er researchers\, we have implemented our method i
n C++ and provide an interface with R software as
an R package BDgraph\, which is freely available
online. \;
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR