BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Geometric extremal graphical models - Jennifer Wadsworth (Lancaste
 r)
DTSTART:20251010T130000Z
DTEND:20251010T140000Z
UID:TALK237490@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:A geometric representation for multivariate extremes\, based o
 n the shapes of scaled sample clouds in light-tailed margins and their so-
 called limit sets\, has recently been shown to connect several existing ex
 tremal dependence concepts. Furthermore\, this geometric representation pr
 ovides a natural way to describe complex extremal dependence structures\, 
 which more established approaches to multivariate extremes do not represen
 t well. These attractive properties have led to recent work that exploits 
 the geometric approach as a foundation for statistical modelling\, which h
 as been demonstrated in relatively low dimensions thus far. For higher dim
 ensional modelling\, we require principled simplifications of the model st
 ructure. We will introduce the concept of geometric extremal graphical mod
 els\, and outline some theoretical results based on block graphs. On the p
 ractical side\, we will demonstrate some initial results employing these i
 deas to model joint river flows in the northwest of England. Based on join
 t works with Ioannis Papastathopoulos\, and Kristina Grolmusova and Thordi
 s Thorarinsdottir.
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR