Paper: arXiv 2611.25881

Authors: Marco van der Burgt

Abstract

The Vasicek distribution assumes an infinite portfolio of identical obligors and forms the basis of the Basel III/IV Risk Weighted Assets (RWA) calculation. We derive the central moments of the loss distribution for a finite number (n) of obligors. We use these moments to calculate percentiles of the fraction (x) of defaulted obligors, via the Cornish-Fisher expansion and via a weighting method. Both methods are based on scaling the standard deviation in x by a capital multiple. The weighting method assumes that the capital multiple is linear in the central moments and leads to an Extended Vasicek distribution. We apply this distribution in calculating an adjusted RWA for small portfolios. Simulations show that this adjustment is only relevant for small portfolios of less than 100 obligors and that the Extended Vasicek distribution allows analytical risk capital calculations for portfolios with different rating grades, even when the number of obligors per rating grade is small.

Complexity vs Empirical Score

  • Math Complexity: 8.0/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a strong theoretical derivation of central moments for finite credit portfolios, extending a foundational model. It then applies these derivations to practical RWA calculations, supported by simulations, demonstrating both mathematical depth and empirical relevance. The methodology is clearly articulated, though full reproducibility would benefit from explicit code/data sharing.

Research Flowchart

  flowchart TD
    A["Research Goal: Extend Vasicek for Finite Portfolios & Small N"] --> B{"Key Methodology: Deriving Central Moments of Loss Distribution"};
    B --> C{"Computational Process: Cornish-Fisher Expansion & Weighting Method"};
    C --> D{"Data/Inputs: Portfolio Size (n), Default Probabilities, Capital Multiple"};
    D --> E["Outcome: Extended Vasicek Distribution"];
    E --> F["Key Findings: Adjusted RWA for Small Portfolios (n<100), Analytical Risk Capital for Multi-Grade Portfolios"];

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 an Extended Vasicek distribution that uses a linear weighting of central moments to calculate percentiles of loss for finite homogeneous credit portfolios.

Source sentence
The weighting method assumes that the capital multiple is linear in the central moments and leads to an Extended Vasicek distribution. We refer to the second method, giving Equation (21) for percentiles, as the Extended Vasicek distribution.

Assumes 3

  • The credit portfolio is homogeneous with equal weights for all exposures. model
    Source sentence
    The credit portfolio is also assumed homogeneous, with equal weights for all exposures.
  • The analysis derives central moments for a finite number (n) of obligors. model
    Source sentence
    We derive the central moments of the loss distribution for a finite number (n) of obligors.
  • The capital multiple is assumed linear in the central moments (1/n‑weighting). model
    Source sentence
    The weighting method assumes that the capital multiple is linear in the central moments and leads to an Extended Vasicek distribution.

Data

  • Monte Carlo simulations of loss distribution (500,000 runs) synthetic unknown
    Source sentence
    The important simulation outcome is the 99.9% percentile (𝑥99.9% 𝑠𝑖𝑚), based on the Harrell-Davis estimator (F.E. Harrell, 1982). a Loss Distribution of a Finite Homogeneous Credit Portfolio – M.J. van der Burgt 19 The figure shows that the percentiles 𝑥99.9% 𝐸𝑉 of the Extended Vasicek distribution are considerably closer to the 𝑥99.9% 𝑠𝑖𝑚 than the percentiles 𝑥99.9% 𝑉 of the original Vasicek distribution for a

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • The adjustment to Basel III/IV RWA is only relevant for small portfolios of less than 100 obligors. theoretical
    Source sentence
    Simulations show that this adjustment is only relevant for small portfolios of less than 100 obligors and that the Extended Vasicek distribution allows analytical risk capital calculations for portfolios with different rating grades, even when the number of obligors per rating grade is small.
  • The Extended Vasicek distribution yields percentiles that closely match the 500,000 simulation results, especially for small numbers of obligors. statistical figure from elsewhere in the paper
    Source sentence
    a Loss Distribution of a Finite Homogeneous Credit Portfolio – M.J. van der Burgt 19 The figure shows that the percentiles 𝑥99.9% 𝐸𝑉 of the Extended Vasicek distribution are considerably closer to the 𝑥99.9% 𝑠𝑖𝑚 than the percentiles 𝑥99.9% 𝑉 of the original Vasicek distribution for a small number of obligors. Overall, the percentiles based on the Extended Vasicek distribution in Equation (21) agree fairly well with
  • The Cornish‑Fisher method gives too low percentiles and capital multiples for small probability‑of‑default values. negative
    Source sentence
    The Cornish-Fisher method gives too low percentiles and capital multiples for small PDs.
  • When the number of obligors exceeds 100, the Extended Vasichk distribution converges to the original Vasicek distribution. theoretical
    Source sentence
    When 𝑛> 100, the Extended Vasicek distribution approaches the original Vasicek distribution, valid for 𝑛→∞.
  • Basel III/IV RWA calculations are sufficiently prudent for finite portfolios unless the portfolio is unrealistically small (n ≤ 20). theoretical
    Source sentence
    First, comparing the Basel III/IV RWA calculation with our alternative RWA, based on the Extended Vasicek distribution, shows that the Basel III/IV calculation includes enough prudency for finite portfolios, unless the portfolios are unrealistically small (n ≤ 20).

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 25 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-20 · same card as JSON · A language model chose these items: check the source sentence before relying on one.