TY  - JOUR
A1  - Schmidt, Patrick
A1  - Katzfuß, Matthias
A1  - Gneiting, Tilmann
T1  - Interpretation of point forecasts with unknown directive
T2  - Journal of applied econometrics
N2  - Point forecasts can be interpreted as functionals (i.e., point summaries) of predictive distributions. We extend methodology for the identification of the functional based on time series of point forecasts and associated realizations. Focusing on state-dependent quantiles and expectiles, we provide a generalized method of moments estimator for the functional, along with tests of optimality under general joint hypotheses of functional relationships and information bases. Our tests are more flexible, and in simulations better calibrated and more powerful than existing solutions. In empirical examples, economic growth forecasts and model output for precipitation are indicative of overstatement in anticipation of extreme events.
KW  - expectile
KW  - identifying moment conditions
KW  - information set
KW  - loss function
KW  - optimality of point forecasts
KW  - quantile
Y1  - 2021
UR  - http://publikationen.ub.uni-frankfurt.de/frontdoor/index/index/docId/63932
UR  - https://nbn-resolving.org/urn:nbn:de:hebis:30:3-639327
SN  - 1099-1255
N1  - The work of Patrick Schmidt, Tilmann Gneiting, and Stephan Hemri was partially funded by the Klaus Tschira Foundation and by the European Union Seventh Framework Programme under Grant 290976. Tilmann Gneiting also is grateful for travel support and encouragement through the ECMWF Fellowship Programme. Matthias Katzfuss was partially supported by US National Science Foundation (NSF) Grant DMS–1521676 and NSF CAREER Grant DMS–1654083. Further, the authors are grateful to the co-editor, three anonymous referees, Werner Ehm, Tobias Fissler, Alexander Glas, Peter Knippertz, Fabian Krüger, Barbara Rossi, Michael Scheuerer, and Peter Vogel for a wealth of constructive and insightful comments. Open access funding enabled and organized by Project DEAL.
N1  - This article has been awarded Open Data Badge for making publicly available the digitally-shareable data necessary to reproduce the reported results. Data is available at https://urldefense.com/v3/__http://qed.econ.queensu.ca/jae/datasets/schmidt001/_;!!N11eV2iwtfs!6j4_zQDGzI8sbkUY6gwZsPRSGkClzhGKs1LQDigJ-b40vBc_o23m3njqu-KcTRGz$.
VL  - 36
IS  - 6
SP  - 728
EP  - 743
PB  - Wiley
CY  - Chichester [u.a.]
ER  -