BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Exchangeability - Peter Orbanz (University of Cambridge)
DTSTART:20110317T140000Z
DTEND:20110317T153000Z
UID:TALK29126@talks.cam.ac.uk
CONTACT:Konstantina Palla
DESCRIPTION:Data is called exchangeable if the probability of a sample dep
 ends only on the values that are observed\, but not on the order in which 
 they occur. This simple  assumption has a rather surprising consequence: A
 n\nexchangeable distribution can be decomposed completely into a shared\, 
 underlying random "pattern" and a component representing independent "rand
 omness" in each observation. This result is de Finetti's theorem. I will d
 evote roughly the first half of the talk to the theorem and its immediate 
 implications for Bayesian statistics. In the remaining time\, I hope to pr
 ovide a glimpse of more recent results and of the fascinating picture whic
 h emerges with them\, which relates symmetry and invariance principles to 
 the design of probability models.\n
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR