Paper: arXiv 2510.05809
Authors: Martin Aichele, Igor Cialenco, Damian Jelito, Marcin Pitera
Abstract
We develop a statistical framework for risk estimation, inspired by the axiomatic theory of risk measures. Coherent risk estimators – functionals of P&L samples inheriting the economic properties of risk measures – are defined and characterized through robust representations linked to \(L\)-estimators. The framework provides a canonical methodology for constructing estimators with sound financial and statistical properties, unifying risk measure theory, principles for capital adequacy, and practical statistical challenges in market risk. A numerical study illustrates the approach, focusing on expected shortfall estimation under both i.i.d. and overlapping samples relevant for regulatory FRTB model applications.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 4.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper is mathematically dense, featuring advanced robust representation theory, duality arguments, and characterizations via L-estimators. While it includes a numerical study, it lacks code, detailed backtesting frameworks, or implementation-heavy data analysis, focusing primarily on theoretical construction and illustration rather than empirical validation.
Research Flowchart
flowchart TD A["Research Goal:<br>Develop a statistical framework for coherent risk estimation"] --> B["Methodology:<br>Axiomatic risk theory & robust representations"] B --> C["Inputs:<br>Price/return samples P&L"] C --> D["Computation:<br>Construct & apply coherent risk estimators<br>e.g., L-estimator for Expected Shortfall"] D --> E["Analysis:<br>Backtesting & simulation under i.i.d.<br>and overlapping samples"] E --> F["Outcomes:<br>Unified methodology for market risk;<br>Improved capital adequacy for FRTB"]
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 a statistical framework for coherent risk estimators (CREs) with a robust representation and structural classification of all CREs.
Source sentence
The key novel contribution of this paper is the development of the robust representation and structural classification of all CREs, see Theorem 4.1.
Assumes 3
- The underlying sample X is an i.i.d. sequence. statistical
Source sentence
For simplicity, throughout this work we assume that X is an i.i.d. sequence.
- Some results assume the CRE is comonotonic. model
Source sentence
assuming additionally that a CRE is comonotonic, we show that it can be represented as a single L-estimator; see Theorem 4.10.
- Consistency results are conjectured to extend under ergodic stationarity (e.g., AR, MA, GARCH). statistical
Source sentence
We conjecture that the definition of consistency, together with Proposition 5.3, can be established under suitable ergodic stationarity assumptions, see (Hayashi, 2011, Section 2).
Data
- Fama–French dataset market-prices unknown
Source sentence
Finally, we checked the subadditivity violation for the empirical VaR estimator using the Fama–French dataset.
- Simulated data synthetic unknown
Source sentence
Numerical illustrations based on simulated and market data demonstrate that coherence of a risk measure does not necessarily carry over to its estimators.
Kenneth French Data Library
Costs
Not applicable
Finds 5
- Subadditivity was violated in 22.4% of the 4,950 portfolio pairs for the empirical VaR estimator on the Fama–French dataset. negative
Source sentence
Subadditivity was violated in 22.4% of the total 4950 pairs.
- For ρ = 0.9, the GPD plug‑in ES estimator violated subadditivity in more than 10% of runs, with 8% of violations exceeding a 10% excess over the sum of individual ES estimates. negative
Source sentence
For ρ = 0.9, the subadditivity is violated in more than 10% of the runs. Moreover, for 8% of the violations, estimated ES for x + x′ exceeded the sum of individual ES estimates by more than 10%.
- Estimator #3 provides the most conservative risk estimates and the best exceedance‑frequency control, but at the cost of a positive bias and higher scaled RMSE. performance
Source sentence
Estimator #3, which produces more conservative risk estimates and achieves improved control of the exceedance frequency. The improved coverage of estimator #3, however, comes at the cost of a positive (conservative) bias and a deterioration in scaled RMSE.
- Estimator #2 has the smallest absolute statistical bias across scenarios, while estimator #1 shows superior variability‑adjusted error. performance
Source sentence
Estimator #2, in turn, exhibits the smallest absolute statistical bias throughout the considered scenarios.
- Parametric plug‑in estimators for Gaussian distributions may fail monotonicity, and semi‑parametric ES estimators for the GPD are not subadditive with positive probability. theoretical
Source sentence
parametric plug‑in estimators for Gaussian distributions may fail monotonicity; semi‑parametric ES estimators for the GPD are not subadditive with positive probability.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 47 related papers). Coverage is partial, so this is neither replication nor refutation.
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 (53)
- Econometrics (2011)
- Exact distribution of the max/min of two G aussian random variables (2008)
- Higher order elicitability and Osbands’ principle (2016)
- CBUAE Rulebook (2023) — link
- EBA report results from the 2024 M arket R isk benchmarking exercise – Part 1 - IMA (2025)
- Market risk, Supervisory Statement, SS13/13 (2020)
- Supervisory Policy Manual, MR-1,V.1 – 15.03.2024 (2024)
- A Universal Stress Scenario Approach for Capitalising Non-modellable Risk Factors Under the FRTB (2021)
and 45 more
- Direct versus iterated multiperiod Value-at-Risk forecasts (2023)
- Error of VaR by overlapping intervals (2009)
- Forecasting: principles and practice (2018)
- A comparison of expected shortfall estimation models (2015)
- Estimation methods for expected shortfall (2014)
- Nonparametric estimation of expected shortfall (2008)
- Backtesting VaR and expectiles with realized scores (2019)
- Robustness in the optimization of risk measures (2022)
- Sample Quantiles in Statistical Packages (1996)
- Dynamic assessment indices (2016)
- On law invariant coherent risk measures (2001)
- ECB guide to internal models, market risk chapters (July 2025) (2025) — link
- Convex optimization (2004)
- Non-standard analysis for coherent risk estimation: hyperfinite representations, discrete Kusuoka formulae, and plug-in asymptotics — → 2602.00784
- A strong law for linear functions of order statistics (1980)
- Some limit behavior for linear combinations of order statistics (2021)
- Strong laws for $L$-and $U$-statistics (1996)
- Approximation theorems of mathematical statistics (1980)
- Some characterizations of strong laws for linear functions of order statistics (1982)
- Nonasymptotic convergence rates for the plug-in estimation of risk measures (2023)
- On the coherence of expected shortfall (2002)
- Optimal nonparametric estimation of the expected shortfall risk — → 2405.00357
- Asymptotic distribution of law-invariant risk functionals (2010)
- A first course in order statistics (2008)
- Distributions determined by cutting a simplex with hyperplanes (1968)
- Asymptotic Statistics (1998)
- Order Statistics (2003)
- Fair estimation of capital risk allocation (2020)
- Conditional value-at-risk for general loss distributions (2002)
- Uniform limit theorems for functions of order statistics (2009)
- Central limit theorems for law-invariant coherent risk measures (2012)
- A modified functional delta method and its application to the estimation of risk functionals (2010)
- Risk management with expectiles (2017)
- Final report – Draft RTS on the assessment methodology under which competent authorities verify an institution’s compliance with the internal model approach as per Article 325az(8) of Regulation (EU) No 575/2013 (Capital Requirements Regulation 2 - CRR2) (2023)
- Commission Delegated Regulation (EU) 2024/397 of 20 October 2023 supplementing Regulation (EU) No 575/2013 of the European Parliament and of the Council with regard to regulatory technical standards on the calculation of the stress scenario risk measure (2024)
- Commission Delegated Regulation (EU) 2024/1085 of 13 March 2024 supplementing Regulation (EU) No 575/2013 of the European Parliament and of the Council with regard to regulatory technical standards on the assessment methodology under which competent authorities verify an institution’s compliance with the requirements to use internal models for market risk (2024)
- Moments of the generalized hyperbolic distribution (2011)
- Time consistency of dynamic risk measures and dynamic performance measures generated by distortion functions — → 2309.02570
- A forecast comparison of volatility models: does anything beat a GARCH (1, 1)? (2005)
- Modeling the conditional distribution of financial returns with asymmetric tails (2020)
- Backtesting value-at-risk and expected shortfall in the presence of estimation error (2023)
- Regression-based expected shortfall backtesting (2022)
- Fama- F rench Data Library (2026)
- Fat tails, VaR and subadditivity (2013)
- Quantitative Risk Management: Concepts, Techniques and Tools (2015)
3 of 53 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.