BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Asymptotic Inference for Score Decompositions - Timo Dimitriadis (
 Goethe-Universität Frankfurt)
DTSTART:20250605T083000Z
DTEND:20250605T085000Z
UID:TALK230848@talks.cam.ac.uk
DESCRIPTION:\nEconomic and statistical forecasts are regularly assessed by
  strictly consistent scoring functions such as the mean squared error for 
 mean\, or the tick loss for quantile forecasts. While scoring functions al
 low for statistically meaningful rankings of overall forecast quality\, th
 ey are silent about the specific deficiencies of the forecasts. We develop
  asymptotic inference for recently proposed score decompositions into misc
 alibration\, discrimination and uncertainty terms\, which allows for hypot
 hesis tests and the construction of confidence intervals for the miscalibr
 ation and discrimination terms. These methods deliver more detailed insigh
 ts into forecast performances in applications to mean forecasts for inflat
 ion rates and volatility forecasts in risk management.\n&nbsp\;\nCo-author
 : Marius Puke\, University of Hohenheim\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR