Higher order elicitability and Osband's principle
From MaRDI portal
Publication:309736
DOI10.1214/16-AOS1439zbMATH Open1355.62006arXiv1503.08123OpenAlexW846844103MaRDI QIDQ309736FDOQ309736
Authors: Tobias Fissler, Johanna F. Ziegel 
Publication date: 7 September 2016
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: A statistical functional, such as the mean or the median, is called elicitable if there is a scoring function or loss function such that the correct forecast of the functional is the unique minimizer of the expected score. Such scoring functions are called strictly consistent for the functional. The elicitability of a functional opens the possibility to compare competing forecasts and to rank them in terms of their realized scores. In this paper, we explore the notion of elicitability for multi-dimensional functionals and give both necessary and sufficient conditions for strictly consistent scoring functions. We cover the case of functionals with elicitable components, but we also show that one-dimensional functionals that are not elicitable can be a component of a higher order elicitable functional. In the case of the variance this is a known result. However, an important result of this paper is that spectral risk measures with a spectral measure with finite support are jointly elicitable if one adds the `correct' quantiles. A direct consequence of applied interest is that the pair (Value at Risk, Expected Shortfall) is jointly elicitable under mild conditions that are usually fulfilled in risk management applications.
Full work available at URL: https://arxiv.org/abs/1503.08123
Recommendations
- Making and evaluating point forecasts
- Order-sensitivity and equivariance of scoring functions
- On the indirect elicitability of the mode and modal interval
- The mode functional is not elicitable
- Of quantiles and expectiles: consistent scoring functions, Choquet representations and forecast rankings. With discussion and authors' reply
consistencyexpected shortfalldecision theoryelicitabilityscoring functionsvalue at riskscoring rulesforecasterOsband's principlepoint forecastsproprietyspectral risk measuresunique minimizer
Applications of statistics to actuarial sciences and financial mathematics (62P05) Decision theory (91B06) Minimax procedures in statistical decision theory (62C20) General considerations in statistical decision theory (62C05)
Cites Work
- Coherent dispersion criteria for optimal experimental design
- Strictly Proper Scoring Rules, Prediction, and Estimation
- Coherent measures of risk
- Higher order elicitability and Osband's principle
- Quantile regression.
- DISTRIBUTION‐INVARIANT RISK MEASURES, INFORMATION, AND DYNAMIC CONSISTENCY
- Robust Estimation of a Location Parameter
- Making and evaluating point forecasts
- Asymmetric Least Squares Estimation and Testing
- Coherence and elicitability
- Elicitation of Personal Probabilities and Expectations
- Title not available (Why is that?)
- Quasi-Hadamard differentiability of general risk functionals and its application
- Risk measures with the CxLS property
- Law invariant risk measures have the Fatou property
- Robustness and sensitivity analysis of risk measurement procedures
- On elicitable risk measures
- Comparative and qualitative robustness for law-invariant risk measures
- On the Optimality of Conditional Expectation as a Bregman Predictor
- Elicitable distortion risk measures: a concise proof
- Verification of internal risk measure estimates
- On the Measurement of Economic Tail Risk
- Title not available (Why is that?)
- External Risk Measures and Basel Accords
- The mode functional is not elicitable
- Qualitative and infinitesimal robustness of tail-dependent statistical functionals
Cited In (79)
- Conditionally Elicitable Dynamic Risk Measures for Deep Reinforcement Learning
- On the Measurement of Economic Tail Risk
- Making and evaluating point forecasts
- Time-varying parameters realized GARCH models for tracking attenuation bias in volatility dynamics
- Uniform calibration tests for forecasting systems with small lead time
- Dynamic semiparametric models for expected shortfall (and value-at-risk)
- Semi-parametric Bayesian tail risk forecasting incorporating realized measures of volatility
- Quantile Evaluation, Sensitivity to Bracketing, and Sharing Business Payoffs
- Dynamic expected shortfall: a spectral decomposition of tail risk across time horizons
- Dynamic large financial networks \textit{via} conditional expected shortfalls
- Scenario-based risk evaluation
- Robustness in the Optimization of Risk Measures
- Forecast dominance testing via sign randomization
- Multivariate geometric expectiles
- Risks in emerging markets equities: time-varying versus spatial risk analysis
- On the indirect elicitability of the mode and modal interval
- Simple factor realized stochastic volatility models
- Joint generalized quantile and conditional tail expectation regression for insurance risk analysis
- XVA analysis from the balance sheet
- Backtesting VaR and expectiles with realized scores
- An energy-based measure for long-run horizon risk quantification
- A Theory for Measures of Tail Risk
- Robust and Pareto optimality of insurance contracts
- Multivariate Shortfall Risk Allocation and Systemic Risk
- Scoring interval forecasts: equal-tailed, shortest, and modal interval
- Forecast evaluation of quantiles, prediction intervals, and other set-valued functionals
- Isotonic regression for elicitable functionals and their Bayes risk
- Quantile-Based Risk Sharing
- A joint quantile and expected shortfall regression framework
- Why scoring functions cannot assess tail properties
- Bayesian realized-GARCH models for financial tail risk forecasting incorporating the two-sided Weibull distribution
- Extremiles: A New Perspective on Asymmetric Least Squares
- Bayes risk, elicitability, and the Expected Shortfall
- Correction to: ``Higher order elicitability and Osband's principle
- On the elicitability of range value at risk
- Elicitability and identifiability of set-valued measures of systemic risk
- Higher order elicitability and Osband's principle
- On the properties of the Lambda value at risk: robustness, elicitability and consistency
- Properization: constructing proper scoring rules via Bayes acts
- Deep quantile and deep composite triplet regression
- S&P 500 volatility, volatility regimes, and economic uncertainty
- Risk Aversion in Regulatory Capital Principles
- Multistep quantile forecasts for supply chain and logistics operations: bootstrapping, the GARCH model and quantile regression based approaches
- A semi-parametric conditional autoregressive joint value-at-risk and expected shortfall modeling framework incorporating realized measures
- Title not available (Why is that?)
- Verification of internal risk measure estimates
- PELVE: probability equivalent level of VaR and ES
- Estimating value-at-risk and expected shortfall using the intraday low and range data
- Performance measurement with expectiles
- Optimal operational service levels in vendor managed inventory contracts -- an exact approach
- DISTORTION RISKMETRICS ON GENERAL SPACES
- Backtesting expected shortfall and beyond
- Multivariate extensions of expectiles risk measures
- Asymptotic stability of empirical processes and related functionals
- Range-based risk measures and their applications
- Regression diagnostics meets forecast evaluation: conditional calibration, reliability diagrams, and coefficient of determination
- Better the devil you know: improved forecasts from imperfect models
- Predicting the Global Minimum Variance Portfolio
- Estimation of complier expected shortfall treatment effects with a binary instrumental variable
- Assessing the accuracy of exponentially weighted moving average models for Value-at-Risk and Expected Shortfall of crypto portfolios
- Elicitability of Instance and Object Ranking
- On joint marginal expected shortfall and associated contribution risk measures
- Multiple measures realized GARCH models
- Modelling additive extremile regression by iteratively penalized least asymmetric weighted squares and gradient descent boosting
- Two-step online estimation and inference for expected shortfall regression with streaming data
- Backtestability and the ridge backtest
- Comparing Predictive Accuracy in the Presence of a Loss Function Shape Parameter
- Realized Quantiles*
- Backtesting Systemic Risk Forecasts Using Multi-Objective Elicitability
- Powerful Backtests for Historical Simulation Expected Shortfall Models
- Sequential testing for elicitable functionals via supermartingales
- Sensitivity measures based on scoring functions
- Review of statistical approaches for modeling high-frequency trading data
- Estimation and backtesting of risk measures with emphasis on distortion risk measures
- Loss function-based change point detection in risk measures
- Uncertainty Comparison Between Value-at-Risk and Expected Shortfall
- Risk quantization by magnitude and propensity
- Inference for joint quantile and expected shortfall regression
- Forecasting Value at Risk and Expected Shortfall Using a Semiparametric Approach Based on the Asymmetric Laplace Distribution
This page was built for publication: Higher order elicitability and Osband's principle
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q309736)
