Sampling of probability measures in the convex order by Wasserstein projection

DOI10.1214/19-AIHP1014zbMATH Open1469.60065OpenAlexW2916196187MaRDI QIDQ2227463FDOQ2227463

Authors Aurélien Alfonsi, Jacopo Corbetta, Benjamin Jourdain 

Publication date 15 February 2021

Published in Annales de l'Institut Henri Poincaré. Probabilités et Statistiques (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1709.05287

zbMATH Keywords

linear programmingWasserstein distanceconvex ordermartingale optimal transportsampling techniques

Mathematics Subject Classification ID

Derivative securities (option pricing, hedging, etc.) (91G20) Variational problems in a geometric measure-theoretic setting (49Q20) Inequalities; stochastic orderings (60E15) Special problems of linear programming (transportation, multi-index, data envelopment analysis, etc.) (90C08)

Abstract: In this paper, for $mu$ and $u$ two probability measures on $mathbb{R}^d$ with finite moments of order $hoge1$, we define the respective projections for the $W_{h}o$-Wasserstein distance of $mu$ and $u$ on the sets of probability measures dominated by $u$ and of probability measures larger than $mu$ in the convex order. The $W_2$-projection of $mu$ can be easily computed when $mu$ and $u$ have finite support by solving a quadratic optimization problem with linear constraints. In dimension $d=1$, Gozlan et al.~(2018) have shown that the projections do not depend on $ho$. We explicit their quantile functions in terms of those of $mu$ and $u$. The motivation is the design of sampling techniques preserving the convex order in order to approximate Martingale Optimal Transport problems by using linear programming solvers. We prove convergence of the Wasserstein projection based sampling methods as the sample sizes tend to infinity and illustrate them by numerical experiments.

Recommendations

• scientific article; zbMATH DE number 7733427
• Sampling in a weighted Sobolev space
• Sampling inequalities in Sobolev spaces
• Sampled Gromov Wasserstein
• Stochastic approximation versus sample average approximation for Wasserstein barycenters
• Sample out-of-sample inference based on Wasserstein distance
• Sharp Wasserstein estimates for integral sampling and Lorentz summability of transport densities
• Log-Sobolev inequalities and sampling from log-concave distributions
• A class of Wasserstein metrics for probability distributions

Cites Work

• scientific article; zbMATH DE number 3778409 (Why is no real title available?)
• scientific article; zbMATH DE number 605729 (Why is no real title available?)
• scientific article; zbMATH DE number 2152346 (Why is no real title available?)
• A new class of costs for optimal transport planning
• A user's guide to measure theoretic probability
• An explicit martingale version of the one-dimensional Brenier theorem
• Characterization of a class of weak transport-entropy inequalities on the line
• Complete lattices of probability measures with applications to martingale theory
• Computational methods for martingale optimal transport problems
• Existence, duality, and cyclical monotonicity for weak transport costs
• Generalized Iterative Scaling for Log-Linear Models
• Intrinsic stationarity for vector quantization: foundation of dual quantization
• Iterative Bregman projections for regularized transportation problems
• Kantorovich duality for general transport costs and applications
• Model-independent bounds for option prices -- a mass transport approach
• On a mixture of Brenier and Strassen theorems
• On a problem of optimal transport under marginal martingale constraints
• On the rate of convergence in Wasserstein distance of the empirical measure
• Optimal quadratic quantization for numerics: the Gaussian case
• Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
• Sampling of one-dimensional probability measures in the convex order and computation of robust option price bounds
• Stability of the shadow projection and the left-curtain coupling
• Stochastic Order Relations and Lattices of Probability Measures
• Stochastic and convex orders and lattices of probability measures, with a martingale interpretation
• The Existence of Probability Measures with Given Marginals

Cited In (21)

• Convex order, quantization and monotone approximations of ARCH models
• A new family of one dimensional martingale couplings
• Continuity of the martingale optimal transport problem on the real line
• Squared quadratic Wasserstein distance: optimal couplings and Lions differentiability
• Instability of martingale optimal transport in dimension \(\mathrm{d}\ge 2\)
• Shadow couplings
• Martingale Wasserstein inequality for probability measures in the convex order
• Approximation of martingale couplings on the line in the adapted weak topology
• Functional convex order for the scaled McKean-Vlasov processes
• An optimal transport-based characterization of convex order
• Backward and forward Wasserstein projections in stochastic order
• Optimal dual quantizers of \(1 D\log \)-concave distributions: uniqueness and Lloyd like algorithm
• Sampling s-Concave Functions: The Limit of Convexity Based Isoperimetry
• Weak Optimal Transport with Unnormalized Kernels
• Lipschitz continuity of the Wasserstein projections in the convex order on the line
• Monotone convex order for the McKean-Vlasov processes
• Weak transport for non‐convex costs and model‐independence in a fixed‐income market
• Convex ordering for stochastic Volterra equations and their Euler schemes
• Stability of martingale optimal transport and weak optimal transport
• Quantization and martingale couplings
• Sampling of one-dimensional probability measures in the convex order and computation of robust option price bounds

This page was built for publication: Sampling of probability measures in the convex order by Wasserstein projection

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2227463)