Answers are grounded in this paper and cite their sources.
This article is concerned with point forecasting of a real-valued random variable with a general Lebesgue density. Answering a question of Gneiting (2011), it is shown that the mode is not elicitable, or, in other words, that it is impossible to find a loss or scoring function under which the mode is the Bayes predictor.
No takes yet. Share an insight, caveat, or question.
Claudio Heinrich (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: