Paper: arXiv 2411.14646

Authors: Xia Han, Liyuan Lin, Hao Wang, Ruodu Wang

Abstract

A diversification quotient (DQ) quantifies diversification in stochastic portfolio models based on a family of risk measures. We study DQ based on expectiles, offering a useful alternative to conventional risk measures such as Value-at-Risk (VaR) and Expected Shortfall (ES). The expectile-based DQ admits simple formulas and has a natural connection to the Omega ratio. Moreover, the expectile-based DQ is not affected by small-sample issues faced by VaR-based or ES-based DQ due to the scarcity of tail data. The expectile-based DQ exhibits pseudo-convexity in portfolio weights, allowing gradient descent algorithms for portfolio selection. We show that the corresponding optimization problem can be efficiently solved using linear programming techniques in real-data applications. Explicit formulas for DQ based on expectiles are also derived for elliptical and multivariate regularly varying distribution models. Our findings enhance the understanding of the DQ’s role in financial risk management and highlight its potential to improve portfolio construction strategies.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 6.0/10 (capped from 6.5: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 6.5)

Why this score: The paper introduces advanced mathematical concepts like expectiles, pseudo-convexity, and explicit formulas under specific distribution models, indicating high mathematical density. It supports claims with empirical evidence, including backtests with real data, computational efficiency comparisons, and simulations, demonstrating substantial implementation and data rigor.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Propose Expectile-based DQ"] --> B["Methodology: Theoretical Derivation"]
  B --> C["Data: Real-world<br>Portfolio Datasets"]
  C --> D["Computation:<br>Linear Programming"]
  D --> E["Key Outcomes"]
  
  subgraph E ["Key Findings"]
      E1["Simple Formulas<br>& Omega Ratio Link"]
      E2["Robust to<br>Small-Sample Issues"]
      E3["Pseudo-convexity &<br>Gradient Descent Feasibility"]
  end

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 diversification quotient based on expectiles as an alternative to VaR and ES, with simple formulas and a connection to the Omega ratio.

Source sentence
We study DQ based on expectiles, offering a useful alternative to conventional risk measures such as Value-at-Risk (VaR) and Expected Shortfall (ES).

Assumes 5

  • The level α is in (0, 1/2). statistical
    Source sentence
    For a given α ∈(0, 1/2), DQex α has the alternative formula
  • The loss vector X follows an elliptically distributed model. model
    Source sentence
    A random vector X is elliptically distributed if its characteristic function can be written as
  • Data are iid (or satisfy an ergodicity condition) for empirical estimation. data
    Source sentence
    Assume we have data X(1), . . . , X(N) sampled from X satisfying some ergodicity condition (being iid would be sufficient).
  • Portfolio weights sum to one and are non‑negative. execution
    Source sentence
    selecting portfolio weights w = (w1, . . . , wn) ∈∆n to minimize DQ based expectiles, where ∆n := {w ∈[0, 1]n : w1 + · · · + wn = 1} .
Show 1 more
  • No transaction costs are incurred in the back‑test. execution
    Source sentence
    assuming no transaction costs.

Data

Backtest sample: 2016-01 – 2023-12 · 40 US stocks · equities · monthly

  • S&P 500 sector portfolios (40 stocks) market-prices unknown
    Jan 2014 to Dec 2023 · 40 stocks from S&P 500 sectors
    Source sentence
    We use data from January 2, 2014, to December 28, 2023 to build up the portfolio, with the risk‑free rate r = 2.13%
  • Synthetic equicorrelated normal data synthetic unknown
    Source sentence
    An illustration of a portfolio vector from an equicorrelated normal distribution is …

Recognised sources: Yahoo Finance

Costs

Not modeled A trading or execution result is reported without costs.

Source sentence
assuming no transaction costs.

Finds 3

Reported performance (Minimizing DQex α (expectile-based diversification quotient)): annual return 10.2%

  • Minimizing the expectile‑based DQ yields the highest Sharpe ratio and lower volatility compared to strategies based on VaR, ES or maximizing the Omega ratio. performance
    Source sentence
    We can see from Table 3 that the strategy based on minimizing DQex α , with the highest Sharp ratio, is quite competitive compared with other strategies. In particular, the annualized volatility for minimizing DQex α is relatively lower than that of maximizing the Omega ratio.
  • Expectile‑based DQ increases steadily with α, while VaR‑ and ES‑based DQ remain zero for small α due to scarce tail data. statistical
    Source sentence
    When α is less than 0.1, both DQVaR α and DQES α are zero, and as α increases, they provide no further useful information. However, the expectile‑based DQ steadily increases with α for α ∈(0, 0.1).
  • The expectile‑based DQ exhibits pseudo‑convexity, allowing efficient gradient‑descent optimization and linear programming solutions. theoretical
    Source sentence
    The expectile‑based DQ exhibits pseudo‑convexity in portfolio weights, allowing gradient descent algorithms for portfolio selection. We show that the corresponding optimization problem can be efficiently solved using linear programming techniques in real‑data applications.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 17 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 (36)

  1. Disappointment concordance and duet expectiles (2024) — link
  2. An extreme worst-case risk measure by expectile (2024) — link
  3. Diversification quotients based on VaR and ES — → 2301.03517
  4. Distributionally robust reinsurance with expectile (2023) — link
  5. Diversification Quotients: Quantifying Diversification via Risk Measures (2022) — link
  6. Extreme M-quantiles as risk measures: From $L^{1}$ to $L^{p}$ optimization (2019) — link
  7. Critical Analysis of the New Basel Minimum Capital Requirements for Market Risk (2017) — link
  8. Extended omega ratio optimization for risk-averse investors (2017) — link
and 28 more
  1. Optimal reinsurance with expectile (2016) — link
  2. Linear programming models based on Omega ratio for the Enhanced Index Tracking Problem (2016) — link
  3. Diversification limit of quantiles under dependence uncertainty (2016) — link
  4. Statistical inference for expectile-based risk measures (2016) — link
  5. Risk concentration based on Expectiles for extreme risks under FGM copula (2015) — link
  6. Elicitation complexity of statistical properties (2015) — link
  7. Aggregation-robustness and model uncertainty of regulatory risk measures (2015) — link
  8. Optimizing the Omega ratio using linear programming (2014) — link
  9. Worst-case robust Omega ratio (2014) — link
  10. An Academic Response to Basel 3.5 (2014) — link
  11. Diversification in heavy-tailed portfolios: properties and pitfalls (2012) — link
  12. On optimal portfolio diversification with respect to extreme risks (2010) — link
  13. Multivariate extremes and the aggregation of dependent risks: examples and counter-examples (2009) — link
  14. Capital Allocation to Business Units and Sub-Portfolios: the Euler Principle (2007) — link
  15. Gain, Loss, and Asset Pricing (2000) — link
  16. Coherent Measures of Risk (1999) — link
  17. A Theory of Disappointment Aversion (1991) — link
  18. Asymmetric Least Squares Estimation and Testing (1987) — link
  19. Probability: Theory and Examples (Vol. 49) (2019)
  20. Joint Mixability (2016) — link
  21. Quantitative Risk Management: Concepts, Techniques and Tools Revised edition (2015)
  22. Generalized Quantiles as Risk Measures (2014) — link
  23. Mathematical Risk Analysis (2013) — link
  24. Making and evaluating point forecast (2011)
  25. Toward maximum diversification (2008)
  26. Robust profit opportunities in risky financial portfolios (2005)
  27. A Universal Performance Measure (2002)
  28. PSEUDO-CONVEX FUNCTIONS (1965) — link

1 of 36 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.