Partial comonotonicity and distortion riskmetrics

Paper: arXiv 2506.07472

Authors: Muqiao Huang

Abstract

We establish a connection between dependence structures and subclasses of distortion riskmetrics under which the latter are additive. A new notion of positive dependence, called partial comonotonicity, is developed, which nests the existing concepts of comonotonicity and single-point concentration. For two random variables, being comonotonic with a third one does not imply that they are comonotonic; instead, this defines an instance of partial comonotonicity. Any specific instance of partial comonotonicity uniquely characterizes a class of distortion riskmetrics through additivity under this dependence structure. An implication of this result is the characterization of the Expected Shortfall using single-point concentration.

Complexity vs Empirical Score

• Math Complexity: 8.5/10
• Empirical Rigor: 1.5/10
• Quadrant: Lab Rats — theoretically deep, empirically untested

Why this score: The paper provides a rigorous mathematical framework introducing ‘partial comonotonicity’ to generalize dependence structures and characterize distortion riskmetrics via additivity. While highly strong on theoretical derivations and proofs, it is purely theoretical without any empirical data analysis or numerical backtesting.

Research Flowchart

  flowchart TD
    A[Goal: Characterize additivity of distortion riskmetrics via dependence structures] --> B[Inputs: Random variables & riskmetrics]
    B --> C[Develop Partial Comonotonicity framework]
    C --> D[Analyze comonotonicity with a 3rd variable & single-point concentration]
    D --> E[Outcome: Establish unique mapping between dependence & riskmetric classes]
    E --> F[Outcome: Characterize Expected Shortfall using single-point concentration]