BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Local Independence Graphs - Niels Richard Hansen (University of Co
 penhagen)
DTSTART:20221104T140000Z
DTEND:20221104T150000Z
UID:TALK182726@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:Conditional local independence\, or Granger non-causality\, is
  a notion of conditional independence among coordinates of a multivariate 
 stochastic process. Local independence graphs can be used to encode such c
 onditional local independencies\, and the abstract and asymmetric independ
 ence models encoded by directed\, and possibly cyclic graphs\, are of intr
 insic interest.\n\nIn the talk I will introduce conditional local independ
 ence and local independence graphs via a classical example of a time homog
 eneous multivariate Markov process with binary coordinates. This will illu
 strate how local independence graphs relate to classical graphical models\
 , such as the Ising model\, but also how they generalize such\nmodels by a
 llowing for an asymmetric dependence over time. I will then show some of t
 he main results we know about local independence graphs\, such as marginal
 ization operations and a characterization of Markov equivalence classes\, 
 and I will outline how local independence graphs can be learned via condit
 ional local independence testing.
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR