BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:New two-sample tests based on adjacency - Hao Chen (University of 
 California\, Davis)
DTSTART:20180322T100000Z
DTEND:20180322T110000Z
UID:TALK102793@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Two-sample tests for multivariate data and non-Euclidean data 
 are widely used in many fields.&nbsp\; We study a nonparametric testing pr
 ocedure that utilizes graphs representing the similarity among observation
 s.&nbsp\; It can be applied to any data types as long as an informative si
 milarity measure on the sample space can be defined.&nbsp\; Existing tests
  based on a similarity graph lack power either for location or for scale a
 lternatives. A new test is proposed that utilizes a common pattern overloo
 ked previously\, and it works for both types of alternatives.&nbsp\; The t
 est exhibits substantial power gains in simulation studies. Its asymptotic
  permutation null distribution is derived and shown to work well under fin
 ite samples\, facilitating its application to large data sets.&nbsp\; Anot
 her new test statistic will also be discussed that addresses the problem o
 f the classic test of the type under unequal sample sizes.&nbsp\; Both tes
 ts are illustrated on an application of comparing networks under different
  conditions.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR