Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2407.13002 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929425393647616 |
|---|---|
| author | Bayraktar, Erhan Norgilas, Dominykas |
| author_facet | Bayraktar, Erhan Norgilas, Dominykas |
| contents | In this article we revisit the weak optimal transport (WOT) problem, introduced by Gozlan, Roberto, Samson and Tetali (2017). We work on the real line, with barycentric cost functions, and as our first result give the following characterization of the set of optimal couplings for two probability measures $μ$ and $ν$: every optimizer couples the left tails of $μ$ and $ν$ using a submartingale, the right tails using a supermartingale, while the central region is coupled using a martingale.
We then consider a constrained optimal transport problem, where admissible transport plans are only those that are optimal for the WOT problem with $L^1$ costs. The constrained problem generalizes the (sub/super-) martingale optimal transport problems, studied by Beiglböck and Juillet (2016), and Nutz and Stebegg (2018) among others.
Finally, we introduce a generalized \textit{shadow measure} and establish its connection to the WOT. This extends and generalizes the results obtained in (sub/super-) martingale settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_13002 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generalizing Super/Sub MOT using weak $L^1$ transport Bayraktar, Erhan Norgilas, Dominykas Probability 60G42, 90C46, 58E30 In this article we revisit the weak optimal transport (WOT) problem, introduced by Gozlan, Roberto, Samson and Tetali (2017). We work on the real line, with barycentric cost functions, and as our first result give the following characterization of the set of optimal couplings for two probability measures $μ$ and $ν$: every optimizer couples the left tails of $μ$ and $ν$ using a submartingale, the right tails using a supermartingale, while the central region is coupled using a martingale. We then consider a constrained optimal transport problem, where admissible transport plans are only those that are optimal for the WOT problem with $L^1$ costs. The constrained problem generalizes the (sub/super-) martingale optimal transport problems, studied by Beiglböck and Juillet (2016), and Nutz and Stebegg (2018) among others. Finally, we introduce a generalized \textit{shadow measure} and establish its connection to the WOT. This extends and generalizes the results obtained in (sub/super-) martingale settings. |
| title | Generalizing Super/Sub MOT using weak $L^1$ transport |
| topic | Probability 60G42, 90C46, 58E30 |
| url | https://arxiv.org/abs/2407.13002 |