BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Score-based calibration testing for multivariate forecast distribu
 tions - Marc-Oliver Pohle (Karlsruhe Institute of Technology (KIT))
DTSTART:20250605T085000Z
DTEND:20250605T091000Z
UID:TALK230851@talks.cam.ac.uk
DESCRIPTION:Malte Kn&uuml\;ppel (Deutsche Bundesbank)\, Fabian Kr&uuml\;ge
 r (Karlsruhe Institute of Technology)\, Marc-Oliver Pohle (Karlsruhe Insti
 tute of Technology\, Heidelberg Institute for Theoretical Studies)\nCalibr
 ation tests based on the probability integral transform (PIT) are routinel
 y used to assess the quality of univariate distributional forecasts. Howev
 er\, PIT-based calibration tests for multivariate distributional forecasts
  face various challenges. We propose two new types of tests based on prope
 r scoring rules\, which overcome these challenges. They arise from a gener
 al framework for calibration testing in the multivariate case\, introduced
  in this work. The new tests have good size and power properties in simula
 tions and solve various problems of existing tests. We apply the tests to 
 forecast distributions for macroeconomic and financial time series data.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR