• Corpus ID: 250264808

A characterisation of convex order using the 2-Wasserstein distance

@inproceedings{Wiesel2022ACO,
  title={A characterisation of convex order using the 2-Wasserstein distance},
  author={Johannes Wiesel and Eric Zhang},
  year={2022}
}
. We give a new characterisation of convex order using the 2-Wasserstein distance W 2 : we show that two probability measures µ and ν on R d with finite second moments are in convex order (i.e. µ (cid:22) c ν ) iff holds for all probability measures ρ on R d with bounded support. Our proof of this result relies on a quantitative bound for the infimum of (cid:82) f dν − (cid:82) f dµ over all 1-Lipschitz functions f , which is obtained through optimal transport duality and Brenier’s theorem. We use… 
2 Citations

Figures from this paper

Martingale Transports and Monge Maps

It is well known that martingale transport plans between marginals µ (cid:54) = ν are never given by Monge maps—with the understanding that the map is over the first marginal µ , or forward in time.

A short proof of the characterisation of convex order using the 2-Wasserstein distance

. We provide a short proof of the intriguing characterisation of the convex order given by Wiesel and Zhang [5].

References

SHOWING 1-10 OF 36 REFERENCES

How to quantise probabilities while preserving their convex order

We introduce an algorithm which, given probabilities µ ≤ cx ν in convex order and defined on a separable Banach space B , constructs finitely-supported approximations µ n → µ, ν n → ν which are in

Squared quadratic Wasserstein distance: optimal couplings and Lions differentiability

In this paper, we remark that any optimal coupling for the quadratic Wasserstein distance W22(μ,ν) between two probability measures μ and ν with finite second order moments on ℝd is the composition

Polar Factorization and Monotone Rearrangement of Vector-Valued Functions

Given a probability space (X, p) and a bounded domain R in Rd equipped with the Lebesgue measure 1 . I (normalized so that 10 I = I ), it is shown (under additional technical assumptions on X and Q)

A note on stochastic dominance, uniform integrability and lattice properties

  • M. Nendel
  • Mathematics
    Bulletin of the London Mathematical Society
  • 2020
In this work, we discuss completeness for the lattice orders of first and second order stochastic dominance. The main results state that both first‐ and second‐order stochastic dominance induce

Irreducible convex paving for decomposition of multidimensional martingale transport plans

Martingale transport plans on the line are known from Beiglbock & Juillet to have an irreducible decomposition on a (at most) countable union of intervals. We provide an extension of this

On a mixture of Brenier and Strassen Theorems

We give a characterization of optimal transport plans for a variant of the usual quadratic transport cost introduced in Gozlan, Roberto, Samson and Tetali (J. Funct. Anal. 273 (2017) 3327–3405).

Complete Duality for Martingale Optimal Transport on the Line

We study the optimal transport between two probability measures on the real line, where the transport plans are laws of one-step martingales. A quasi-sure formulation of the dual problem is

Computational methods for martingale optimal transport problems

It is proved that the MOT value can be approximated using linear programming (LP) problems which result from a discretisation of the marginal distributions combined with a suitable relaxation of the martingale constraint.