The mode functional is not elicitable
成果类型：
Article
署名作者：
Heinrich, C.
署名单位：
Ruprecht Karls University Heidelberg
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.1093/biomet/ast048
发表日期：
2014
页码：
245251
关键词：
摘要：
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.
来源URL：