Abstract: Point forecasts can be interpreted as functionals (i.e., point summaries) of predictive distributions. We consider the situation of unknown directives and discuss how to estimate the functional based on time series data 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 and more specific hypotheses. In empirical examples, the gross domestic product (GDP) Greenbook forecasts of the US Federal Reserve and model output for precipitation at London from the European Centre for Medium-Range Weather Forecasts (ECMWF) are indicative of overshooting in anticipation of extreme events. Joint work with Patrick Schmidt (Heidelberg Institute for Theoretical Studies and Goethe University Frankfurt) and Matthias Katzfuss (Texas A&M University). Personal webpage: Professor Tilmann Gneiting
Professor Tilmann Gneiting is head of the Computational Statistics (CST) group at the Heidelberg Institute for Theoretical Studies (HITS gGmbH) and Professor of Computational Statistics at the Karlsruhe Institute of Technology (KIT) in Germany.