BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Asymptotics for In-Sample Density Forecasting - Enno Mammen\, Univ
 ersity of Mannheim
DTSTART:20140207T160000Z
DTEND:20140207T170000Z
UID:TALK50324@talks.cam.ac.uk
CONTACT:20082
DESCRIPTION:In in-sample density forecasting the density of observations i
 s estimated in regions where the density is not observed. Identification o
 f the density in such regions is guaranteed by structural assumptions on t
 he \ndensity that allows exact extrapolation. In this talk the structural 
 assumption is made that the density is \na product of one-dimensional func
 tions. The theory is quite general in assuming the shape of the region \nw
 here the density is observed. Such models naturally arise when the time po
 int of an observation can be \nwritten as the sum of two terms (e.g. onset
  and incubation period 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 studie
 s. Kernel smoothing estimators are proposed that are based \non backfittin
 g.  Full asymptotic theory is derived for them. The talk reports  on joint
  work with Young K. Lee\, \nMaria Dolores Martinez-Miranda\, Jens P. Niels
 en and Byeong U. Park.
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR