Forecasting Conditional Probabilities of Binary Outcomes under Misspecification

Graham Elliott; Dalia Ghanem; Fabian Krüger

doi:10.1162/rest_a_00564

Forecasting Conditional Probabilities of Binary Outcomes under Misspecification

Graham Elliott¹; Dalia Ghanem²; Fabian Krüger³

¹ University of California San Diego · ² University of California, Davis, and Giannini Foundation · ³ Heidelberg Institute for Theoretical Studies (HITS)

The Review of Economics and Statistics 2016 open access

Abstract We consider constructing probability forecasts from a parametric binary choice model under a large family of loss functions (“scoring rules”). Scoring rules are weighted averages over the utilities that heterogeneous decision makers derive from a publicly announced forecast (Schervish, 1989). Using analytical and numerical examples, we illustrate howdifferent scoring rules yield asymptotically identical results if the model is correctly specified. Under misspecification, the choice of scoring rule may be inconsequential under restrictive symmetry conditions on the data-generating process. If these conditions are violated, typically the choice of a scoring rule favors some decision makers over others.

DOI: 10.1162/rest_a_00564
Volume: 98 (4)
Pages: 742-755
Language: en
Export: BibTeX
Sources: openalex crossref