Proper Calibeating

Dean P. Foster and Sergiu Hart



(Acrobat PDF files)


Abstract

The classic concept of "calibrated forecasts," and its more recent refinement of "calibeating," are defined with respect to the standard quadratic scoring rule. We extend these notions to the class of proper scoring rules (for which the best forecast is the true distribution), and define proper calibration and proper calibeating by requiring the errors to converge to zero uniformly over all proper Lipschitz scoring rules. We first establish that calibration always implies proper calibration, whereas calibeating need not imply proper calibeating. Second, we show how to guarantee proper calibeating, as well as continuous proper calibeating by a deterministic procedure, and proper multicalibeating.







   


Last modified:
© Sergiu Hart