The directional optimal transport

The directional optimal transport
复制标题

DOI:
10.1214/21-aap1712
复制
发表时间:
2020-02
期刊:
The Annals of Applied Probability
影响因子:
--
通讯作者:
Marcel Nutz;Ruodu Wang
Marcel Nutz;Ruodu Wang
中科院分区:
其他
文献类型:
--
作者:
Marcel Nutz;Ruodu Wang

文献摘要

相似文献

我们引入一个约束最优运输问题,其中起点 $x$ 只能运输到目的地 $y\geq x$。我们的统计动机是描述当效果单调时给定边际的治疗效果 $Y-X$ 方差的尖锐上限,或 $Y\geq X$。因此,我们关注超模成本(或子模奖励),并引入一个耦合 $P_{*}$,它对于所有此类成本都是最优的,并产生锐界。这种耦合允许多种表征——几何、序论、通过 cdf 和传输内核作为最佳传输——解释其结构并暗示有用的边界。当第一个边际是无原子的时,$P_{*}$集中在两个可以用边际描述的映射的图上,第二个映射由于绑定约束而产生。
We introduce a constrained optimal transport problem where origins $x$ can only be transported to destinations $y\geq x$. Our statistical motivation is to describe the sharp upper bound for the variance of the treatment effect $Y-X$ given marginals when the effect is monotone, or $Y\geq X$. We thus focus on supermodular costs (or submodular rewards) and introduce a coupling $P_{*}$ that is optimal for all such costs and yields the sharp bound. This coupling admits manifold characterizations---geometric, order-theoretic, as optimal transport, through the cdf, and via the transport kernel---that explain its structure and imply useful bounds. When the first marginal is atomless, $P_{*}$ is concentrated on the graphs of two maps which can be described in terms of the marginals, the second map arising due to the binding constraint.