Paper: arXiv 2509.19151
Authors: Fengnan Deng, Anand N. Vidyashankar, Jeffrey F. Collamore
Abstract
We obtain sharp large deviation estimates for exceedance probabilities in dependent triangular array threshold models with a diverging number of latent factors. The prefactors quantify how latent-factor dependence and tail geometry enter at leading order, yielding three regimes: Gaussian or exponential-power tails produce polylogarithmic refinements of the Bahadur-Rao \(n^{"-1/2"}\) law; regularly varying tails yield index-driven polynomial scaling; and bounded-support (endpoint) cases lead to an \(n^{"-3/2"}\) prefactor. We derive these results through Laplace-Olver asymptotics for exponential integrals and conditional Bahadur-Rao estimates for the triangular arrays. Using these estimates, we establish a Gibbs conditioning principle in total variation: conditioned on a large exceedance event, the default indicators become asymptotically i.i.d., and the loss-given-default distribution is exponentially tilted (with the boundary case handled by an endpoint analysis). As illustrations, we obtain second-order approximations for Value-at-Risk and Expected Shortfall, clarifying when portfolios operate in the genuine large-deviation regime. The results provide a transferable set of techniques-localization, curvature, and tilt identification-for sharp rare-event analysis in dependent threshold systems.
Complexity vs Empirical Score
- Math Complexity: 9.5/10
- Empirical Rigor: 3.2/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper is highly theoretical, featuring advanced probability theory (large deviations, Laplace-Olver asymptotics, triangular arrays) and dense mathematical derivations without any code, backtesting, or empirical implementation. Its rigor is purely theoretical, aimed at establishing general asymptotic results rather than data-driven validation.
Research Flowchart
flowchart TD
A["Research Goal<br>Sharp rare-event analysis in<br>dependent credit threshold models"] --> B["Methodology<br>Laplace-Olver asymptotics<br>& Conditional Bahadur-Rao"]
B --> C["Computational Processes<br>Localization, curvature,<br>tilt identification"]
C --> D{"Data/Input: Tail Geometry?"}
D --> E["Regime 1: Gaussian/Exp-Power<br>Polylogarithmic refinement"]
D --> F["Regime 2: Regularly Varying<br>Polynomial scaling"]
D --> G["Regime 3: Bounded-Support<br>n^{'-3/2} prefactor"]
E --> H["Findings: Gibbs Conditioning<br>Defaults asymptotically i.i.d.<br>LGD exponentially tilted"]
F --> H
G --> H
H --> I["Outcomes: Risk Metrics<br>VaR/ES 2nd-order approx.<br>Large-deviation regime identification"]
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 sharp large deviation estimates for exceedance probabilities in dependent triangular array threshold models with many latent factors, and establishes a Gibbs conditioning principle in total variation for large-loss events.
Source sentence
We obtain sharp large deviation estimates for exceedance probabilities in dependent triangular array threshold models with a diverging number of latent factors. Using these estimates, we establish a Gibbs conditioning principle in total variation: conditioned on a large exceedance event, the default indicators become asymptotically i.i.d., and the loss-given-default distribution is exponentially tilted (with the
Assumes 6
- Normalization of factor loadings: sum of squared loadings plus squared idiosyncratic weight equals one for all n. model
Source sentence
We assume throughout the article that the nonnegative constants {a(n) j : j = 1, . . . , kn} and b(n) have been normalized such that kn X j=1 (a(n) j )2 + (b(n))2 = 1, for all n; (A1) and we further assume that b := lim n→∞b(n) ∈(0, 1).
- Definition of aggregate common factor Zn as the weighted sum of latent factors. model
Source sentence
(A2) For ease of notation, denote the aggregate common factors by Zn := a(n) 1 Z(n) 1 + · · · + a(n) kn Z(n) kn , n ∈Z+.
- Limit of idiosyncratic weight b(n) exists and lies in (0,1). model
Source sentence
We assume throughout the article that the nonnegative constants {a(n) j : j = 1, . . . , kn} and b(n) have been normalized such that kn X j=1 (a(n) j )2 + (b(n))2 = 1, for all n; (A1) and we further assume that b := lim n→∞b(n) ∈(0, 1).
- Loss sizes U(n)i are i.i.d. and independent of default indicators and latent factors. model
Source sentence
Next, let U (n) i denote the loss incurred from the ith loan (the “loss given default”, or LGD), and assume that {U (n) i : i = 1, . . . , n} is an i.i.d. sequence (possibly dependent on n). We assume that the loss size is independent of the event of default; thus, {U (n) i : i = 1, . . . , n} may be taken to be independent of {Z(n) j : j = 1, . . . , kn} and of {ϵ(n) i : i = 1, . . . , n} (see Appendix F, Remark
Show 2 more
- Log‑smooth tail conditions for the idiosyncratic factor ε (Gumbel‑type) and the common factor Z. statistical
Source sentence
(Log-smooth tail condition for ϵ (Gumbel-type)). (Log-smooth left-tail condition for Zn).
- Balance and window interiority conditions (W1‑W3) ensuring appropriate scaling of the tail region. statistical
Source sentence
(Balance condition). There exist a sequence Mn ∈(an, bn), where an, bn →∞, such that R1n := n Cx b(n) fϵ( v+Mn b(n) ) rZn(Mn) →1 and R2n := nCxfϵ( v+Mn b(n) )Q′ ϵ( v+Mn b(n) ) (b(n)rZn(Mn))2 = Θ(1). (W1).
Data
- Synthetic Gaussian ε and Gaussian Z synthetic unknown
Source sentence
Case 1: Gaussian ϵ and Gaussian Z. Table E.1 reports VaR and ES estimates for the setting ϵ ∼N(0, 1) and Z ∼N(0, 1). n = 10 n = 50 n = 100 n = 500 n = 1000 VaR α = 0.95 0.565 0.522 0.514 0.505 0.503 α = 0.99 0.630 0.550 0.534 0.513 0.509 α = 0.999 0.711 0.589 0.561 0.526 0.518 ES α = 0.95 0.692 0.597 0.572 0.537 0.528 α = 0.99 0.692 0.583 0.558 0.526 0.518 α = 0.999 0.745 0.607 0.574 0.532 0.522 Table E.1: VaRα and
- Synthetic Gaussian ε and Pareto Z synthetic unknown
Source sentence
Gaussian Z. Case 2: Gaussian ϵ and Pareto Z. Table E.2 presents the results when ϵ ∼N(0, 1) and Z ∼ Pareto(xm = 1, α = 2). Compared with the Gaussian case, the heavy-tailed distribution of Z yields systematically larger values of both VaRα and ESα, reflecting the heavier tail risk. n = 10 n = 50 n = 100 n = 500 n = 1000 VaR α = 0.95 0.664 0.560 0.539 0.514 0.510 α = 0.99 0.718 0.587 0.559 0.524 0.516 α = 0.999 0.777
Costs
Not applicable
Finds 5
- Sharp large deviation prefactors exhibit three regimes: polylogarithmic refinements for Gaussian or exponential‑power tails, polynomial scaling for regularly varying tails, and an n^{-3/2} prefactor for bounded‑support tails. theoretical
Source sentence
The prefactors quantify how latent-factor dependence and tail geometry enter at leading order, yielding three regimes: Gaussian or exponential-power tails produce polylogarithmic refinements of the Bahadur-Rao $n^{-1/2}$ law; regularly varying tails yield index-driven polynomial scaling; and bounded-support (endpoint) cases lead to an $n^{-3/2}$ prefactor.
- Conditioned on a large loss event, default indicators become asymptotically i.i.d. and the loss‑given‑default distribution is exponentially tilted. theoretical
Source sentence
Using these estimates, we establish a Gibbs conditioning principle in total variation: conditioned on a large exceedance event, the default indicators become asymptotically i.i.d., and the loss-given-default distribution is exponentially tilted (with the boundary case handled by an endpoint analysis).
- Second‑order expansions for Value‑at‑Risk and Expected Shortfall are derived, showing when these risk measures lie in the genuine large‑deviation regime. theoretical
Source sentence
As illustrations, we obtain second-order approximations for Value-at-Risk and Expected Shortfall, clarifying when portfolios operate in the genuine large-deviation regime. Setting xα,n = µU + yn and applying a Taylor expansion up to third order around µU, we obtain Λ∗ U(xα,n) = Λ∗ U(µU + yn) = Λ∗ U(µU) + (Λ∗ U)′(µU)yn + 1 2(Λ∗ U)′′(µU)y2 n + O(y3 n), (E.1) 47 Noticing that Λ∗ U(µU) = 0, (Λ∗ U)′(µU) = 0, and (Λ∗
- Numerical experiments confirm that heavy‑tailed common factors (Pareto Z) produce systematically larger VaR and ES values compared to Gaussian factors. performance
Source sentence
Compared with the Gaussian case, the heavy-tailed distribution of Z yields systematically larger values of both VaRα and ESα, reflecting the heavier tail risk. n = 10 n = 50 n = 100 n = 500 n = 1000 VaR α = 0.95 0.664 0.560 0.539 0.514 0.510 α = 0.99 0.718 0.587 0.559 0.524 0.516 α = 0.999 0.777 0.619 0.582 0.535 0.524 ES α = 0.95 0.712 0.587 0.561 0.526 0.518 α = 0.99 0.752 0.606 0.573 0.531 0.521 α = 0.999 0.801
- The Gibbs conditioning principle holds in total variation, identifying the tilted limit law and demonstrating disappearance of dependence under conditioning. theoretical
Source sentence
We also established a Gibbs conditioning principle in total variation for the conditional law under rare-loss events, identifying the tilted limit law and clarifying when dependence disappears at the conditioning scale.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 22 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 (24)
- Higher-order asymptotic expansions for Laplace’s integral and their error estimates (2025) — link
- Refined behaviour of a conditioned random walk in the large deviations regime (2024) — link
- Sharp Probability Tail Estimates for Portfolio Credit Risk (2022) — link
- A transient Cramér–Lundberg model with applications to credit risk (2020) — link
- Geometric sharp large deviations for random projections of ℓpn spheres and balls (2020) — link
- A conditional strong large deviation result and a functional central limit theorem for the rate function (2014) — link
- Default clustering in large portfolios: Typical events. (2011) — link
- Law of large numbers and large deviations for dependent risks (2009) — link
and 16 more
- Portfolio Credit Risk with Extremal Dependence: Asymptotic Analysis and Efficient Simulation (2008) — link
- LARGE DEVIATIONS IN MULTIFACTOR PORTFOLIO CREDIT RISK (2007) — link
- Credit Risk: Pricing, Measurement, and Management (2005) — link
- Credit Risk Modeling: Theory and Applications (2004)
- Π-regular variation (1981) — link
- Asymptotics and Special Functions (1974) — link
- On the Pricing of Corporate Debt: The Risk Structure of Interest Rates (1974) — link
- On the Oscillations of Sums of Independent Random Variables (1970) — link
- CONVERGENCE OF PROBABILITY MEASURES (1970) — link
- On Deviations of the Sample Mean (1960) — link
- Quantitative Risk Management: Concepts, Techniques and Tools Revised edition (2015)
- Extreme value theory : an introduction (2006) — link
- Large portfolio losses (2004) — link
- THE DISTRIBUTION OF LOAN PORTFOLIO VALUE (2002)
- Probability theory: Independence, interchangeability, martingales (1978) — link
- iv) Under additional condition (L1) , then for any fixed z , as n → ∞ , n | Λ ∗ U ( n ) ( x ) − Λ ∗ U ( x ) |→ 0 , n | θ x,n − θ x |→ 0 , and n | Λ ∗ n ( x ; z ) − Λ ∗ ( x ; z ) |→ 0
0 of 24 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.