BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multiple change point estimation based on moving sum statistics  -
  Claudia Kirch (Magdeburg)
DTSTART:20180511T150000Z
DTEND:20180511T160000Z
UID:TALK103837@talks.cam.ac.uk
CONTACT:Quentin Berthet
DESCRIPTION:Using the example of changes in the mean we introduce change p
 oint estimators for multiple changes based on moving sum statistics. For a
  given bandwidth we prove consistency of these estimators as well as their
  asymptotic limit distribution. In many practical situations a single band
 width is not sufficient to detect all change points\, but instead several 
 bandwidths need to be used. We discuss extensions based on information cri
 teria to combine this multiscale information in a consistent way. Finally\
 , extensions to a very general setting based on estimating functions are d
 iscussed. These extensions allow for more robust methods in the mean chang
 e situation as well as for more complex change point problems such as chan
 ges in the time series structure. 
LOCATION:MR12
END:VEVENT
END:VCALENDAR