Robust forecast aggregation
Robust forecast aggregation
复制标题
DOI：
10.1073/pnas.1813934115
复制
发表时间：
2018-12-26
影响因子：
11.1
通讯作者：
Smorodinsky, Rann
中科院分区：
文献类型：
--
作者：
Arieli, Itai;Babichenko, Yakov;Smorodinsky, Rann
接触不同证据的贝叶斯专家经常做出相互矛盾的概率预测。一个聚合器，不知道底层模型，用它来计算他或她自己的预测。我们使用评分规则和遗憾的概念，提出一个自然的方式来评估一个聚合方案。我们专注于一个二进制状态空间，并构建低遗憾聚合方案时，只有两个专家，要么是布莱克威尔有序或条件独立同分布（i.i.d.）信号.相反，如果有许多专家有条件i.i.d.信号，则没有方案（渐近地）比（0.5，0.5）预测更好。
Bayesian experts who are exposed to different evidence often make contradictory probabilistic forecasts. An aggregator, ignorant of the underlying model, uses this to calculate his or her own forecast. We use the notions of scoring rules and regret to propose a natural way to evaluate an aggregation scheme. We focus on a binary state space and construct low regret aggregation schemes whenever there are only two experts that either are Blackwell-ordered or receive conditionally independent and identically distributed (i.i.d.) signals. In contrast, if there are many experts with conditionally i.i.d. signals, then no scheme performs (asymptotically) better than a (0.5, 0.5) forecast.