BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Discrete Statistical Modeling using Chain Event Graphs - Jim Smith
 \, Warwick
DTSTART:20090227T160000Z
DTEND:20090227T170000Z
UID:TALK16638@talks.cam.ac.uk
CONTACT:8419
DESCRIPTION: Chain Event Graphs encode a new class of finite discrete\nmod
 els that strictly contains discrete Bayesian Network models and\ntheir con
 text specific generalizations as a very special case. They\nprovide a part
 icularly powerful graphical framework for eliciting\,\nquerying\, encoding
 \, performing inferences and estimating highly\nasymmetric models in an ef
 ficient and transparent way. Such model\nclasses arise naturally in both i
 n the social sciences and biology.\nThe class exhibits many of the advanta
 ges of Bayesian Networks. There\nare direct analogues of graphical conditi
 onal independence querying\ntechniques. The framework supports conjugate i
 nference with complete\ndata and hence efficient exact search algorithms o
 ver the model class.\nFurthermore\, like the Bayesian Network\, the class 
 encodes algebraic\nconstraints on a class of polynomials and so it can be 
 mapped into its\nown associated an albeit typically inhomogeneous algebrai
 c\nparametrization. Finally\, being constructed from an event tree\, Chain
 \nEvent Graphs admit an excellent framework for expressing causal\nextensi
 ons of this model class. The talk will demonstrate these\nproperties using
  a number of examples.\n
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR