BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian nonparametric approach to testing for dependence betwee
 n random variables  - Sarah Filippi (Oxford)
DTSTART:20160603T150000Z
DTEND:20160603T160000Z
UID:TALK65900@talks.cam.ac.uk
CONTACT:Quentin Berthet
DESCRIPTION:Nonparametric and nonlinear measures of statistical dependence
  between pairs of random variables are important tools in modern data anal
 ysis. In particular the emergence of large data sets can now support the r
 elaxation of linearity assumptions implicit in traditional association sco
 res such as correlation. Here we describe a Bayesian nonparametric procedu
 re that leads to a tractable\, explicit and analytic quantification of the
  relative evidence for dependence vs independence. Our approach uses Polya
  tree priors on the space of probability measures which can then be embedd
 ed within a decision theoretic test for dependence.  Polya tree priors can
  accommodate known uncertainty in the form of the underlying sampling dist
 ribution and provides an explicit posterior probability measure of both de
 pendence and independence. Well known advantages of having an explicit pro
 bability measure include: easy comparison of evidence across different stu
 dies\; encoding prior information\; quantifying changes in dependence acro
 ss different experimental conditions\, and\; the integration of results wi
 thin formal decision analysis.
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberforce Road\, Camb
 ridge.
END:VEVENT
END:VCALENDAR