Wasserstein Distributionally Robust Risk-Sensitive Estimation via Conditional Value-at-Risk
Feras Al Taha, Eilyan Bitar · 2026
Read Wasserstein Distributionally Robust Risk-Sensitive Estimation via Conditional Value-at-Risk by Feras Al Taha, Eilyan Bitar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-18546.txt.
Description
Al Taha and Bitar propose a robust approach to risk-aware estimation from signals. Both signals are random vectors with a joint distribution assumed in a type-2 Wasserstein ball, the ambiguity set. Quality is judged by tail value-at-risk of squared error. Their main result: when the nominal center has finite support, affine estimators best limiting worst-case CVaR can be found via a tractable semidefinite program. On bulk power price outlook with real data, they deliver smaller out-of-sample CVaR of squared error than rivals.
Catalog record
| CREATOR | Feras Al Taha, Eilyan Bitar |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32166 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.18546