BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Change-point tests based on estimating functions - Kirch\, C (Karl
 sruhe Institute of Technology)
DTSTART:20140115T133000Z
DTEND:20140115T140000Z
UID:TALK49923@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Many classical change-point tests are based on cumulative sums
  of estimating functions\, where the most prominent example are quasi maxi
 mum likelihood scores. Examples include testing for changes in the locatio
 n model\, continuous linear and non-linear autoregressive time series as w
 ell as most recently changes in count time series. While classic theory de
 als with offline procedures where the full data set has been observed befo
 re a statistical decision about a change-point is made\, the same principl
 es can be used in sequential testing. The latter has gained some increased
  interest in the last decade\, where initial parameter estimation is based
  on some historic data set with no change-point\, before cumulative sum ch
 arts are used to monitor newly arriving data. In such a setup\, asymptotic
 s are carried out with the size of the historic data set increasing to inf
 inity. In applications such a data set will typically exist as usually at 
 least some data is collected before any reason able statistical inference 
 can be made. In this talk we explain the underlying ideas and extract regu
 larity conditions under which asymptotics both under the null hypothesis a
 s well as alternative can be derived. We will illustrate the usefulness us
 ing different examples that have partly already been discussed in the lite
 rature.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR