arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2109.00464math.STstat.MEstat.TH
收藏

众数相对于单峰分布是否可被 elicit?

Is the mode elicitable relative to unimodal distributions?

Claudio Heinrich-Mertsching, Tobias Fissler

AI总结:

本文证明,即使限制在具有连续 Lebesgue 密度和唯一局部最大值的分布类中,众数仍不可被 elicit,且不可被 identifiy,显著加强了 Heinrich (2014) 的结果。

摘要:

统计泛函被称为可被 elicit 的,如果存在一个损失函数或评分函数,使得在该函数下该泛函是期望意义上的最优点预测。虽然均值和分位数是可被 elicit 的,但 Heinrich (2014) 已证明,若真实分布可以是任意 Lebesgue 密度,则众数无法被 elicit。我们显著加强了这一结果,表明若真实分布是任何具有连续 Lebesgue 密度且具有唯一局部最大值的分布,众数也无法被 elicit。同样地,众数相对于该类别也不可被 identifiy。

英文摘要:

Statistical functionals are called elicitable if there exists a loss or scoring function under which the functional is the optimal point forecast in expectation. While the mean and quantiles are elicitable, it has been shown in Heinrich (2014) that the mode cannot be elicited if the true distribution can follow any Lebesgue density. We strengthen this result substantially, showing that the mode cannot be elicited if the true distribution is any distribution with continuous Lebesgue density and unique local maximum. Likewise, the mode fails to be identifiable relative to this class.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑