BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Conditional Dependence Measure with Applications to Undirected G
 raphical Models - Yang Feng\, Columbia University
DTSTART:20150529T150000Z
DTEND:20150529T160000Z
UID:TALK58615@talks.cam.ac.uk
CONTACT:20082
DESCRIPTION:Measuring conditional dependence is an important topic in stat
 istics with broad applications including graphical models. Under a factor 
 model setting\, a new conditional dependence measure is proposed. The meas
 ure is derived by using distance covariance after adjusting the common obs
 ervable factors or covariates. The corresponding conditional independence 
 test is given with the asymptotic null distribution unveiled. The latter g
 ives a somewhat surprising result: the estimating errors in factor loading
  matrices\, while of root-n order\, do not have material impact on the asy
 mptotic null distribution of the test statistic\, which is also in the roo
 t-n domain. It is also shown that the new test has strict control over the
  asymptotic significance level and can be calculated efficiently. A generi
 c method for building dependency graphs using the new test is elaborated. 
 Numerical results and real data analysis show the superiority of the new m
 ethod.\n
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR