BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Change-point tests based on estimating functions -
Kirch\, C (Karlsruhe Institute of Technology)
DTSTART;TZID=Europe/London:20140115T133000
DTEND;TZID=Europe/London:20140115T140000
UID:TALK49923AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/49923
DESCRIPTION:Many classical change-point tests are based on cum
ulative sums of estimating functions\, where the m
ost prominent example are quasi maximum likelihood
scores. Examples include testing for changes in t
he location model\, continuous linear and non-line
ar autoregressive time series as well as most rece
ntly changes in count time series. While classic t
heory deals with offline procedures where the full
data set has been observed before a statistical d
ecision about a change-point is made\, the same pr
inciples can be used in sequential testing. The la
tter has gained some increased interest in the las
t decade\, where initial parameter estimation is b
ased on some historic data set with no change-poin
t\, before cumulative sum charts are used to monit
or newly arriving data. In such a setup\, asymptot
ics are carried out with the size of the historic
data set increasing to infinity. In applications s
uch a data set will typically exist as usually at
least some data is collected before any reason abl
e statistical inference can be made. In this talk
we explain the underlying ideas and extract regula
rity conditions under which asymptotics both under
the null hypothesis as well as alternative can be
derived. We will illustrate the usefulness using
different examples that have partly already been d
iscussed in the literature.\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR