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:Statistics
SUMMARY:Asymptotics for In-Sample Density Forecasting - En
no Mammen\, University of Mannheim
DTSTART;TZID=Europe/London:20140207T160000
DTEND;TZID=Europe/London:20140207T170000
UID:TALK50324AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/50324
DESCRIPTION:In in-sample density forecasting the density of ob
servations is estimated in regions where the densi
ty is not observed. Identification of the density
in such regions is guaranteed by structural assump
tions on the \ndensity that allows exact extrapola
tion. In this talk the structural assumption is ma
de that the density is \na product of one-dimensio
nal functions. The theory is quite general in assu
ming the shape of the region \nwhere the density i
s observed. Such models naturally arise when the t
ime point of an observation can be \nwritten as th
e sum of two terms (e.g. onset and incubation peri
od of a disease). The developed theory also allows
for a multiplicative factor of seasonal effects.
Seasonal effects are present in many actuarial\, \
nbiostatistical\, econometric and statistical stud
ies. Kernel smoothing estimators are proposed that
are based \non backfitting. Full asymptotic theo
ry is derived for them. The talk reports on joint
work with Young K. Lee\, \nMaria Dolores Martinez
-Miranda\, Jens P. Nielsen and Byeong U. Park.
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberf
orce Road\, Cambridge
CONTACT:
END:VEVENT
END:VCALENDAR