Paper: arXiv 2311.12183
Authors: Silvana M. Pesenti, Steven Vanduffel
Abstract
We employ scoring functions, used in statistics for eliciting risk functionals, as cost functions in the Monge-Kantorovich (MK) optimal transport problem. This gives raise to a rich variety of novel asymmetric MK divergences, which subsume the family of Bregman-Wasserstein divergences. We show that for distributions on the real line, the comonotonic coupling is optimal for the majority of the new divergences. Specifically, we derive the optimal coupling of the MK divergences induced by functionals including the mean, generalised quantiles, expectiles, and shortfall measures. Furthermore, we show that while any elicitable law-invariant coherent risk measure gives raise to infinitely many MK divergences, the comonotonic coupling is simultaneously optimal. The novel MK divergences, which can be efficiently calculated, open an array of applications in robust stochastic optimisation. We derive sharp bounds on distortion risk measures under a Bregman-Wasserstein divergence constraint, and solve for cost-efficient payoffs under benchmark constraints.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 2.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper is mathematically dense, focusing on advanced optimal transport theory, risk measures, and elicitability, with rigorous proofs and derivations; however, it lacks implementation details, backtests, or empirical data, relying instead on theoretical applications in robust optimization.
Research Flowchart
flowchart TD A["Research Goal: Explore MK optimal transport<br>using scoring functions as costs"] --> B["Methodology: Define novel MK divergences<br>from elicitable risk functionals"] B --> C["Analyze Theoretical Properties<br>Optimal couplings, e.g., comonotonic"] C --> D["Computational Process<br>Derive efficient calculations<br>for distributions on the real line"] D --> E["Key Outcomes & Applications<br>1. Generalization of Bregman-Wasserstein divergences<br>2. Robust stochastic optimization bounds<br>3. Cost-efficient payoff design"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes novel asymmetric Monge‑Kantorovich divergences by using statistical scoring functions as transport cost functions, generalising Bregman‑Wasserstein divergences.
Source sentence
We employ scoring functions, used in statistics for eliciting risk functionals, as cost functions in the Monge-Kantorovich (MK) optimal transport problem. This gives raise to a rich variety of novel asymmetric MK divergences, which subsume the family of Bregman-Wasserstein divergences.
Assumes 2
- Scoring functions are normalised as in Assumption 2.2, i.e., strictly consistent on point measures and satisfy S_T(δ_y, y)=0 for all y. statistical
Source sentence
Throughout, we make the following non-restrictive assumptions on the considered scoring functions. Assumption 2.2 (Normalisation of scores).
- The scoring function S is M‑consistent (or strictly M‑consistent) for the elicitable functional T. model
Source sentence
For a given functional T : ¯ M →A and a sub-class M ⊆¯ M, the scoring function S may satisfy the following properties: (i) S is M-consistent for T, if for all F ∈M and for all z ∈A (2.1) Z S �T(F), y dF(y) ≤ Z S(z, y) dF(y) .
Data
No dataset named in the text.
Costs
Not applicable
Finds 3
- For distributions on the real line, the comonotonic coupling is optimal for the majority of the new MK divergences. theoretical
Source sentence
We show that for distributions on the real line, the comonotonic coupling is optimal for the majority of the new divergences. Specifically, we derive the optimal coupling of the MK divergences induced by functionals including the mean, generalised quantiles, expectiles, and shortfall measures.
- All elicitable law‑invariant coherent risk measures generate infinitely many MK divergences, yet the comonotonic coupling remains simultaneously optimal. theoretical
Source sentence
Furthermore, we show that while any elicitable law-invariant coherent risk measure gives raise to infinitely many MK divergences, the comonotonic coupling is simultaneously optimal. Any elicitable risk functional, such as the mean, quantiles, expectiles, admits infinitely many consistent score, and thus gives raise to an infinite family of MK divergences.
- The proposed MK divergences can be efficiently calculated and enable applications such as sharp bounds on distortion risk measures under a Bregman‑Wasserstein constraint and cost‑efficient payoffs under benchmark constraints. theoretical
Source sentence
The novel MK divergences, which can be efficiently calculated, open an array of applications in robust stochastic optimisation. We derive sharp bounds on distortion risk measures under a Bregman-Wasserstein divergence constraint, and solve for cost-efficient payoffs under benchmark constraints.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 10 related papers). Coverage is partial, so this is neither replication nor refutation.
5 papers in the archive : Distributionally Robust Insurance under Bregman-Wasserstein …
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (39)
- Making and Evaluating Point Forecasts (2011) — link
- Higher order elicitability and Osband’s principle (2016) — link
- Mathematical risk analysis (2013)
- Inequalities: theory of majorization and its applications (1979)
- Robust distortion risk measures (2023) — link
- Elicitation of personal probabilities and expectations (1971)
- Forecast evaluation of quantiles, prediction intervals, and other set-valued functionals (2021) — link
- Portfolio optimization within a W asserstein ball (2023) — link
and 31 more
- Asymmetric least squares estimation and testing (1987) — link
- Optimal transport: Old and new (2008)
- Sugli estremi dei momenti delle funzioni di ripartizione doppia (1956)
- Convex majorization with an application to the length of critical paths (1979) — link
- B regman- W asserstein divergence: geometry and applications (2023) — link
- Optimal transport for applied mathematicians (2015)
- The isotonic regression problem and its dual (1972)
- On elicitable risk measures (2015) — link
- Clustering with Bregman divergences. (2005)
- Mirror descent and nonlinear projected subgradient methods for convex optimization (2003)
- On monge’s problem for bregman-like cost functions (2007)
- Bregman monotone optimization algorithms (2003)
- M-quantiles (1988) — link
- Conditional $ L _p$-quantiles and their application to the testing of symmetry in non-parametric regression (1996) — link
- Risk contributions of lambda quantiles (2022)
- On additive premium calculation principles (1974) — link
- Elicitation and identification of properties (2014)
- Sensitivity measures based on scoring functions (2023) — link
- Stochastic finance: an introduction in discrete time (2011) — link
- Law invariant concave utility functions and optimization problems with monotonicity and comonotonicity constraints (2006)
- The optimal payoff for a Yaari investor (2022)
- A note on the quantile formulation (2016)
- Optimal portfolio under state-dependent constraints (2015)
- Portfolio choice via quantiles (2011)
- On the construction of optimal payoffs (2020)
- Eliciting properties of probability distributions (2008) — link
- Probabilistic sensitivity measures as information value (2021)
- On the properties of the L ambda value at risk: robustness, elicitability and consistency (2017) — link
- Sensitivity measures based on scoring functions (2023)
- Mass Transportation Problems: Volume I: Theory (1998)
- Inefficient dynamic portfolio strategies or how to throw away a million dollars in the stock market (1988) — link
0 of 39 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.