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 Methods for Networks - Peter Hoff (Univer
sity of Washington)
DTSTART;TZID=Europe/London:20160725T100000
DTEND;TZID=Europe/London:20160725T110000
UID:TALK66836AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/66836
DESCRIPTION:Statistical analysis of social network data presen
ts many challenges: Realistic models often requir
e a large number of parameters\, yet maximum like
lihood estimates for even the simplest models may
be unstable. Furthermore\, network data often exh
ibit non-standard statistical dependencies\, and m
ost network datasets lack any sort of replication
.

Statistical methods to address th
ese issues have included random effects and laten
t variable models\, and penalized likelihood metho
ds. In this talk I will discuss how these approac
hes fit naturally within a Bayesian framework for
network modeling. Additionally\, we will discuss
how standard Bayesian concepts such as exchangeab
ility play a role in the development and interpret
ation of probability models for networks. Finally
\, some thoughts on the use of Bayesian methods f
or large-scale dynamic networks will be presented.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR