Multi-Trigger Crypto CAT Bonds with On-Chain Settlement: Valuation and Optimal Design

Maochao Xu; Xianyue Li; Yijia Li; Yue Wang

REVIEW 3 major objections 8 minor 32 references

Multi-Trigger Crypto CAT Bonds with On-Chain Settlement: Valuation and Optimal Design

T0 review · 3 major / 8 minor · reviewed 2026-07-09 · glm-5.2

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Crypto catastrophe bonds get arbitrage-free pricing and on-chain settlement

desk verdict First integrated crypto CAT bond framework with on-chain settlement; theory is sound but independence assumption is load-bearing and untested read the letter →

arxiv 2607.06981 v1 pith:ZLSE3TR5 submitted 2026-07-08 stat.AP

Yue Wang , Yijia Li , Maochao Xu , Xianyue Li This is my paper

classification stat.AP

keywords bondsdependencesettlementcryptodesignmulti-triggeron-chainrisk

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper builds a pricing and design framework for catastrophe bonds tailored to cryptocurrency risks—protocol exploits, exchange breaches, and DeFi failures. The central object is a double-trigger structure that fires on either a sudden large single-incident loss (monthly maximum) or a sustained accumulation of moderate losses (cumulative total), with payout multipliers embedded directly into smart contracts for automated settlement. The authors prove that an arbitrage-free pricing measure exists on an enlarged information filtration combining traditional financial market data with oracle-reported crypto-loss states, and that this measure decomposes bond risk into a hedgeable financial component and an irreducible crypto-loss residual. They further show that a sponsor-optimal contract minimizing tail-value-at-risk exists under a dual-measure framework where investor pricing uses the risk-neutral measure Q and sponsor exposure uses the physical measure P. Empirically, the framework is calibrated using generalized extreme value distributions for crypto losses, copulas for trigger dependence, and ARIMA-GARCH with vine copulas for financial discounting factors, yielding simulated price and return distributions across candidate contract designs.

What carries the argument

The load-bearing machinery is the product-form pricing measure Q = Q_F ⊗ P_L ⊗ P_C constructed on the enlarged filtration F_chain = F_fin ∨ F_orc, where Q_F is the minimal martingale measure on tradable financial assets and P_L, P_C are the unchanged physical laws of crypto-loss dynamics and on-chain contract states. Theorem 3.1 establishes that discounted traded asset prices remain Q-martingales under this enlarged filtration, yielding risk-neutral valuation V_0 = E^Q[Σ d(R_k, Y_k, Z_k)/D_k]. Theorem 3.2 decomposes the discounted payoff H into a hedgeable projection H^F onto the self-financing trading gain space and an orthogonal residual H^⊥. Theorem 3.3 proves existence of a sponsor-optim

What would settle it

If crypto-loss dynamics and financial market risk factors exhibit significant dependence during stress periods, the product-form measure Q = Q_F ⊗ P_L ⊗ P_C would not correctly price the bond, and the hedgeable/unhedgeable risk decomposition would be misspecified. The independence assumption is testable: one could examine whether extreme crypto-loss months coincide with financial market dislocations, or whether the 2022-2023 period of simultaneous crypto failures and monetary tightening shows cross-domain dependence that the static copula model fails to capture.

Watch

Extended reading notes

Core claim

The paper's core claim is that crypto-native catastrophe bonds can be priced in an arbitrage-free manner even though crypto-loss risk is non-tradable, by extending the minimal martingale measure from the tradable financial submarket to a product-space measure Q = Q_F ⊗ P_L ⊗ P_C that leaves the crypto-loss distribution unchanged. Under this measure, bond values decompose into a hedgeable financial component and an orthogonal residual risk that investors must bear. The double-trigger mechanism—responding jointly to short-horizon severity and long-horizon accumulation—combined with on-chain smart-contract settlement, produces a risk-transfer instrument that is both statistically calibrated to异

Load-bearing premise

The framework assumes that tradable financial market risk factors (Treasury rates, inflation, SOFR) are statistically independent of crypto-loss dynamics. This factorization is what allows the pricing measure to split cleanly into a financial martingale measure and an unchanged crypto-loss law. If crypto losses and financial conditions become correlated during periods of joint stress—precisely when catastrophe bonds are most needed—the pricing and hedging decomposition may be

Editorial extensions

If this is right

If the framework is correct, crypto-native catastrophe bonds could become a viable mechanism for transferring tens of billions of dollars in protocol-exploit and exchange-breach risk from the crypto ecosystem to capital market investors.
The double-trigger design—combining short-term severity with cumulative accumulation—could serve as a template for other emerging-risk catastrophe bonds where loss dynamics exhibit both sudden shocks and gradual deterioration, such as AI system failures or climate transition risks.
The on-chain settlement architecture, if adopted, would eliminate settlement delays and basis risk that plague traditional catastrophe bonds, potentially compressing the time between trigger event and payout from weeks to hours.
The martingale risk decomposition separating hedgeable financial risk from non-tradable crypto-loss risk provides a template for pricing any insurance-linked security whose underlying peril is uncorrelated with tradable financial assets.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: this is the first paper I've seen that seriously combines arbitrage-free pricing, sponsor-optimal contract design, and smart-contract settlement for crypto catastrophe bonds. The theory is correct under stated assumptions. The main soft spot is the independence assumption between financial and crypto-loss factors, which is load-bearing and empirically untested despite the authors having both data series in hand. Worth a serious referee; the contribution is real if the authors address the gaps in revision. The reader's take is mostly on target. The stress-test concern about independence is the right thing to flag, and it's the issue I'd push hardest in revision. The theorems are fine — Theorem 3.1's product-measure construction is standard, the orthogonality proof in 3.2 is correct, and the existence argument in 3.3 uses conventional compactness/continuity. No issues there. The dual-trigger structure (monthly max + cumulative loss) is a genuine design contribution, and the on-chain settlement architecture, while not formally novel as a concept, is concretely specified and well-motivated for this domain. The empirical work is where things get shaky. The independence assumption is structurally load-bearing: the product measure Q = Q_F ⊗ P_L ⊗ P_C and the martingale property both depend on it. The authors flag it as a 'scale-separation approximation' and acknowledge it may fail during joint stress — which is exactly when CAT bonds trigger. What's frustrating is they have both crypto-loss and financial factor residuals already computed in Section 4.3. A cross-correlation test, especially in the tails, would directly probe the assumption. They don't do it. That's the revision I'd require. The sample sizes are small (31-41 months), and the 2024 'validation' is a single path replay, not out-of-sample testing. The trigger calibration targeting 15% returns introduces partial circularity into the cost analysis. No code or smart contract implementation is shipped, despite the architecture being a selling point. Parameter uncertainty isn't propagated into pricing. These are real but addressable. Who gets value: researchers in ILS, crypto risk, and DeFi insurance. The framework is a legitimate starting point even if the empirical validation needs work. Recommend serious peer review with revision focused on: (1) empirical test of the independence assumption, (2) out-of-sample validation beyond one path, (3) sensitivity analysis showing how pricing changes if independence is relaxed. The theory stands; the empirics need to catch up.

Referee Report

3 major / 8 minor

Summary. This paper proposes a catastrophe (CAT) bond framework tailored to cryptocurrency risks, combining a double-trigger structure (monthly maximum loss and cumulative loss) with on-chain settlement via smart contracts. The theoretical contribution is an arbitrage-free valuation framework under an incomplete-market setting using the minimal martingale measure (MMM) on a product space of financial, crypto-loss, and on-chain contract-state information (Theorems 3.1–3.2), plus a sponsor-optimal contract design existence result under a dual-measure (Q, P) formulation (Theorem 3.3). The empirical implementation models crypto losses via GEV marginals with copula dependence and financial factors via ARIMA-GARCH with vine copulas, feeding Monte Carlo simulation of bond prices, returns, and sponsor tail risk (TVaR). A case study illustrates feasibility using 2024 realized data for an ETH-denominated note.

Significance. The paper addresses a genuinely novel and timely intersection: crypto-native catastrophe risk securitization with on-chain settlement. The theoretical results are mathematically coherent—the product-measure construction in Theorem 3.1 is standard, the martingale orthogonality proof in Theorem 3.2 is correct, and the existence argument in Theorem 3.3 uses standard compactness/continuity tools. The dual-measure design formulation (pricing under Q, sponsor tail risk under P) is a clean conceptual contribution. The statistical calibration is reasonably thorough, with category-specific copula selection and diagnostic checks. The on-chain settlement architecture, while not formally proven, is a plausible and well-motivated operational design. The framework is falsifiable in the sense that simulated price and return distributions can be compared against future realized outcomes, as the 2024 replay attempts.

major comments (3)

Section 3.1, independence assumption: The product-form pricing measure Q = Q_F ⊗ P_L ⊗ P_C and the martingale property in Theorem 3.1 both depend on the assumption that tradable financial risk factors are independent of crypto-loss dynamics. The authors acknowledge this is a 'scale-separation approximation' that 'may not hold exactly during rare periods of joint stress.' This is structurally load-bearing: if crypto losses and financial variables are correlated during tail events—precisely when the CAT bond triggers—the product measure may be misspecified and the hedgeable/unhedgeable decomposition of Theorem 3.2 may not hold. The paper has both crypto-loss and financial data in hand but provides no empirical test of this assumption. A cross-correlation analysis between the crypto-loss series (or their indicators) and the financial factor residuals, even a simple one, would substantially
Table 11 and Section 5, circularity in trigger calibration: The trigger thresholds (Table 9) are calibrated so that 'the simulated mean annualized return is approximately 15%' (Section 5), and the four principal repayment designs L1–L4 are then compared on sponsor TVaR (Table 11). The result that L1 (most aggressive) achieves the lowest TVaR is partly mechanical: more aggressive principal write-down shifts more loss to investors by construction. The comparison would be more informative if the designs were calibrated to a common market-clearing spread (which Table 10 shows they are, but the spread differences are small: 1.07%–1.18%) and if the investor-side tail risk (e.g., probability of principal loss) were reported alongside sponsor TVaR. Without investor-side tail metrics, the trade-off framing is incomplete, and the claim that the framework 'supports economically viable risk transfer
Section 4.3.1, GEV shape parameter estimates: For several fitted models, the shape parameter ξ is not statistically significant (e.g., BSC monthly maximum M2: ξ = −0.0629, SE = 0.1670; Other monthly aggregate M2: ξ = −0.1117, SE = 0.1551). Given that the tail behavior of the loss distribution—and hence the CAT bond trigger probabilities—depends critically on ξ, the paper should discuss the sensitivity of simulated bond prices and tail metrics to the estimated ξ values, or at minimum report confidence intervals for trigger probabilities. The current presentation reports point estimates without uncertainty quantification, which the authors acknowledge as a limitation (Section 7) but which is particularly relevant here given the small sample sizes (31–41 months).

minor comments (8)

Section 5, calibration of trigger thresholds: The procedure for selecting the 15% target return is described only briefly. A clearer statement of why 15% was chosen (market benchmark, calibration to existing cyber ILS spreads, etc.) would help the reader assess whether the resulting cost levels (19–21% of limit in Table 13) are reasonable.
Equation (1): The cash flow function d(R_k, Y_k, Z_k) uses notation that is slightly inconsistent with the surrounding text. The coupon function f_{ξ_k, η_k}(R_k) in Eq. (2) includes the spread s, but Eq. (1) does not explicitly show s as an argument. Clarifying the notation would improve readability.
Section 2, Figure 2: The on-chain settlement architecture is described in prose but the figure is somewhat schematic. Adding labels for the specific smart contract functions (trigger evaluation, coupon distribution, principal settlement) and their interaction with oracle inputs would make the architecture more concrete.
Table 9: The cumulative-loss trigger thresholds (δ^C_1 = 2.37×10^9, δ^C_2 = 2.56×10^9) are very close together relative to their magnitude. The sensitivity of the principal repayment to the exact placement of these thresholds should be discussed, as small changes could shift scenarios between B2 and B3.
Section 4.3.1: The GEV models M1–M4 include time-varying location/scale, but the time variable t is not clearly defined (is it the month index within the sample? calendar time?). Specifying this would aid reproducibility.
Section 6: The 'pure realized return benchmark' of 18.9% for 2024 is based on a single realized path. While the authors note it lies within the simulated distribution, the framing as a 'benchmark' could be misinterpreted. Clarifying that this is an illustrative single-path outcome rather than a validation test would be appropriate.
The abstract states the on-chain design 'eliminates basis risk associated with settlement delays.' This is an overstatement: the design reduces settlement-delay basis risk but does not eliminate model/parameter basis risk (the trigger depends on oracle-reported losses, which may differ from the sponsor's actual economic losses). 'Minimizes' or 'substantially reduces' would be more accurate than 'eliminates.'
References: The paper cites 'arXiv:2607.06981v1' with a date of '8 Jul 2026,' which appears to be a future date. This should be corrected to the actual submission date.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity in core theoretical results; empirical calibration is conventional, not circular

full rationale

The paper's core theoretical results (Theorems 3.1–3.3) are derived from standard assumptions in incomplete-market pricing without circular reasoning. Theorem 3.1 establishes that Q = Q_F ⊗ P_L ⊗ P_C is an equivalent martingale measure on the enlarged filtration F_chain = F_fin ∨ F_orc, using the product-space independence assumption. The proof is self-contained: the martingale property follows because discounted traded asset prices depend only on the Ω_F coordinate, so the additional on-chain information is irrelevant under the product measure. This is a standard extension of the minimal martingale measure to an enlarged filtration, not a self-citation chain or a definitional reduction. Theorem 3.2's risk decomposition (hedgeable vs. unhedgeable components) follows from the L² projection structure and is also independently derived. Theorem 3.3's existence result for sponsor-optimal contracts uses standard compactness and continuity arguments (dominated convergence, dual representation of TVaR) without circularity. On the empirical side, the calibration of trigger thresholds to target a 15% expected return (Section 5) and the subsequent calibration of spreads to satisfy the par-issuance condition (Eq. 5, Table 10) is a standard two-step calibration procedure, not circularity: the thresholds define the contract structure, and the spread is the endogenous price that clears the market under that structure. The 2024 replay (Table 12) serves as an out-of-sample check, producing a realized return of 18.9% that lies within the simulated distribution, which is an independent validation rather than a fitted-then-predicted result. The independence assumption between financial and crypto-loss factors is load-bearing for the product-form measure, but this is a modeling assumption (correctly flagged as an approximation by the authors), not a circularity. No self-citation chain is load-bearing for the central claims. The derivation is self-contained against external benchmarks.

Assumptions & free parameters 9 free parameters · 6 assumptions · 2 invented entities

The paper introduces 9 classes of free parameters (several fitted to data, several chosen by design), 6 axioms (2 standard math, 4 domain assumptions), and 2 invented entities (the crypto CAT bond instrument and the on-chain settlement architecture). The high parameter count is expected for an empirical pricing paper, but the lack of deployed smart contracts or market demand tests means the invented entities have no independent evidence outside the paper's own framework.

free parameters (9)

Trigger thresholds δ^M_1, δ^M_2, δ^M_3 = 1.99e6, 6.68e6, 1.18e7 (ETH)
Monthly maximum loss thresholds calibrated so baseline simulated return ≈ 15%; directly determines trigger activation probabilities and thus bond cash flows.
Trigger thresholds δ^C_1, δ^C_2 = 2.37e9, 2.56e9 (ETH)
Cumulative loss thresholds calibrated alongside monthly max thresholds; determines principal repayment bin assignments.
Coupon spread s = 1.0689%–1.1809% across L1–L4
Calibrated via par-issuance condition V_0(Δ,Γ;s) = F; endogenously determined by the pricing measure and trigger structure.
GEV parameters (μ, σ, ξ) per category = Varies by category and model M1–M4
Location, scale, and shape parameters fitted to log-transformed monthly max and aggregate losses via maximum likelihood.
Copula parameters = Gaussian θ=0.97 (ETH), Gumbel θ=7.32 (BSC), BB1 θ=(1.51, 5.97) (Other)
Fitted to capture dependence between monthly max and aggregate losses; category-specific.
ARIMA-GARCH parameters = See Table 6
Fitted to Treasury, inflation, SOFR series; determines financial discount factor dynamics.
Vine copula parameters = Clayton θ=0.15 (TR-IF), Clayton θ=0.29 (TR-SOFR)
Fitted to standardized residuals of financial series; captures cross-series dependence.
Coupon multiplier matrix Δ(A_ξ B_η) = See Figure 1(a), e.g., 3.0 to 0
Manually specified grid of multipliers; not fitted but chosen by design.
Principal repayment multipliers Γ(B_η) = L1: (1.0, 0.6, 0.0); L4: (1.0, 0.95, 0.90)
Four candidate designs specified for scenario comparison; not fitted but chosen.

assumptions (6)

domain assumption Independence of financial risk factors and crypto-loss dynamics
Section 3.1: 'we assume that the tradable financial risk factors are independent of the on-chain information relevant for contract settlement.' Enables the product-measure construction Q = Q_F ⊗ P_L ⊗ P_C in Theorem 3.1.
standard math Arbitrage-free traded financial submarket admitting a minimal martingale measure
Section 3.1: assumes existence of Q_F ∼ P_F such that discounted traded asset prices are Q_F-martingales. Standard assumption from Föllmer et al. (1990).
domain assumption Oracle-reported losses are the sole loss input to the contract
Section 2: 'the collection {X_{k,j}} constitutes the sole loss input available to the contract for trigger evaluation and settlement.' Assumes oracle integrity and completeness.
domain assumption Static copula dependence structures are time-invariant over the bond horizon
Sections 4.2-4.3: copula parameters are estimated once and held fixed. The authors acknowledge (Section 7) that 'non-stationarity and structural breaks may challenge the validity of static copula assumptions.'
domain assumption Interest-rate risk premia are negligible over the one-year horizon
Section 5: 'we approximate the financial pricing measure for the auxiliary macro factors... by their historically estimated dynamics. Equivalently, we assume that the associated interest-rate risk premia are negligible over this horizon.'
standard math Compactness of the admissible design set D
Theorem 3.3: assumes D is compact and satisfies a uniform lower bound on V(Δ). Enables the existence proof via Weierstrass theorem.

invented entities (2)

Crypto CAT bond (multi-trigger, on-chain settled)
purpose: Risk-transfer instrument for crypto-native catastrophic losses
The instrument itself is the paper's proposal; no market implementation or empirical demand test is provided. The 2024 replay is a backtest on a single path, not an independent market test.
On-chain settlement architecture (4-layer: Data/Pricing, Trigger/Governance, Capital Flow, Bond/Investor)
purpose: Automated trigger evaluation and cash-flow execution via smart contracts
Described conceptually in Section 2 and Figure 2 but not implemented or deployed. No smart contract code, no testnet deployment, no gas cost measurements beyond rough estimates.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Multi-Trigger Crypto CAT Bonds with On-Chain Settlement: Valuation and Optimal Design." pith.science (2026). https://pith.science/paper/ZLSE3TR5

@misc{pith2026260706981,
  author       = {Pith},
  title        = {Pith review of: Multi-Trigger Crypto CAT Bonds with On-Chain Settlement: Valuation and Optimal Design},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZLSE3TR5}},
  note         = {Machine review of arXiv:2607.06981}
}

read the original abstract

Cryptocurrencies have experienced repeated large-scale losses from protocol exploits and exchange breaches, exposing insurers and investors to severe operational risks. This paper develops an equilibrium pricing framework for catastrophe bonds tailored to the cryptocurrency ecosystem. We introduce a double-trigger structure that jointly captures short-term catastrophic shocks and longer-term systemic deterioration. To model the multi-risk environment, we incorporate dual dependence, combining dependence across triggers with multivariate dependence among financial risk factors through vine copulas. Beyond expected prices, we characterize the full distribution of discounted cash flows and return rates, enabling risk-sensitive metrics such as Value-at-Risk and Tail Value-at-Risk. Furthermore, we propose an on-chain settlement architecture where calibrated payout functions are embedded directly into smart contracts. This design eliminates basis risk associated with settlement delays and minimizes the agency costs inherent in traditional intermediation. Our results demonstrate that multi-trigger crypto CAT bonds offer a statistically robust and economically efficient vehicle for transferring systemic digital asset risks to capital markets.

Figures

Figures reproduced from arXiv: 2607.06981 by the authors.

**Figure 2.** Operational architecture for trustless settlement. The automated flow of capital [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗

**Figure 3.** Normal score plots for all categories. 4.1.2 Financial Risks In our bond evaluation, we incorporate financial market risks from three sources: inflation,5 Treasury,6 and SOFR.7 The Treasury and inflation series span from January 2000 to December 2023, while SOFR is available from April 2018 through December 2023. All series are converted to monthly rates to ensure consistent temporal granularity and to facilitate su… view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: Time series of monthly financial risk factors: Treasury, inflation, and SOFR. [PITH_FULL_IMAGE:figures/full_fig_p021_4.png]

**Figure 5.** Figure 5: QQ plots for transformed monthly maximum losses of all categories. [PITH_FULL_IMAGE:figures/full_fig_p026_5.png]

**Figure 6.** Figure 6: QQ plots for transformed monthly aggregate losses of all categories. [PITH_FULL_IMAGE:figures/full_fig_p027_6.png]

**Figure 7.** Figure 7: Contour plots of all blockchain categories, where the blue points are normal scores. [PITH_FULL_IMAGE:figures/full_fig_p028_7.png]

**Figure 8.** Figure 8: QQ plots of fitted financial risk rates. [PITH_FULL_IMAGE:figures/full_fig_p030_8.png]

**Figure 9.** Figure 9: Heatmap of Pearson’s ρ (lower triangle) and Kendall’s τ (upper triangle) correlations among the standardized residuals of Treasury, inflation, and SOFR time series. with ρ = 0.419 and τ = 0.205. The correlation between Treasury and inflation is also nonnegligible (ρ =…

**Figure 10.** Figure 10: Contour plots illustrating the fitted C-vine copula structure linking Treasury, infla [PITH_FULL_IMAGE:figures/full_fig_p032_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

[1]

Barrieu, P., Braun, A., & Makariou, D. (2024). Catastrophe bonds.Handbook of Insurance: Volume I: Third Edition, 169–195. Springer

work page 2024
[2]

W., Gaur, V., & Giesecke, K

Biais, B., Capponi, A., Cong, L. W., Gaur, V., & Giesecke, K. (2023). Advances in blockchain and crypto economics.Management Science, 69(11), 6417–6426

work page 2023
[3]

(2013).Convergence of probability measures

Billingsley, P. (2013).Convergence of probability measures. John Wiley & Sons. Bj¨ ork, T. (2009).Arbitrage theory in continuous time. Oxford university press

work page 2013
[4]

Braun, A., Eling, M., & Jaenicke, C. (2023). Cyber insurance-linked securities.ASTIN Bulletin: The Journal of the IAA, 53(3), 684–705

work page 2023
[5]

(2020).Insurance-linked securities primer.https://content.naic.org/sites/ default/files/capital-markets-primer-linked-securities.pdf

Carelus, J.-B. (2020).Insurance-linked securities primer.https://content.naic.org/sites/ default/files/capital-markets-primer-linked-securities.pdf. [Accessed 10 Septem- ber 2025]. F¨ ollmer, H. & Schweizer, M. (2010). Minimal martingale measure.Encyclopedia of Quantitative Finance, 3, 1200–1204. F¨ ollmer, H., Schweizer, M., et al. (1990).Hedging of cont...

work page 2020
[6]

Gan, J., Tsoukalas, G., & Netessine, S. (2023). Decentralized platforms: Governance, toke- nomics, and ico design.Management Science, 69(11), 6667–6683

work page 2023
[7]

Ghalanos, A. (2020). Introduction to the rugarch package.R vignette

work page 2020
[8]

& Ferreira, A

Haan, L. & Ferreira, A. (2006).Extreme value theory: an introduction. Springer. 41 HKSAR (2021).Estimates of expenditure 2021–22: Financial services and the treasury bureau. https://www.legco.gov.hk/yr20-21/english/fc/fc/w_q/fstb-fs-e.pdf. [Accessed 10 September 2025]

work page 2006

Show all 32 references

[9]

Huang, S., Zhang, J., & Zhu, W. (2024). Storm cat bond: Modeling and valuation.North American Actuarial Journal, 28(4), 718–743

2024
[10]

(2009).Mathematical methods for financial markets

Jeanblanc, M., Yor, M., & Chesney, M. (2009).Mathematical methods for financial markets. Springer Science & Business Media

2009
[11]

(1997).Multivariate models and dependence concepts

Joe, H. (1997).Multivariate models and dependence concepts. Monographs on Statistics and Applied Probability. Chapman & Hall

1997
[12]

(2014).Dependence modeling with copulas

Joe, H. (2014).Dependence modeling with copulas. CRC Press

2014
[13]

John, K., Kogan, L., & Saleh, F. (2023). Smart contracts and decentralized finance.Annual Review of Financial Economics, 15(1), 523–542

2023
[14]

Kolesnikov, O., Markov, A., Smagulov, D., & Solovjovs, S. (2022). Cyber loss distribution fitting: a general framework towards cyber bonds and their pricing models.International Journal of Mathematics and Mathematical Sciences, 2022(1), 7689828

2022
[15]

Kwock, A., Lie, E., Weng, G., & Zhang, R. (2022).Decentralized insurance alternatives: Mar- ket landscape, opportunities and challenges.https://www.soa.org/4aa5c3/globalassets/ assets/files/resources/research-report/2022/decentralized-ins-alt.pdf. [Ac- cessed 10 September 2025...

2022
[16]

Li, H., Liu, H., Tang, Q., & Yuan, Z. (2023). Pricing extreme mortality risk in the wake of the covid-19 pandemic.Insurance: Mathematics and Economics, 108, 84–106

2023
[17]

Li, H. & Su, J. (2024). Mitigating wildfire losses via insurance-linked securities: Modeling and risk management perspectives.Journal of Risk and Insurance, 91(2), 383–414

2024
[18]

N., Xu, M., Zhao, P., & Hu, T

Li, Y., Nguyen, Q. N., Xu, M., Zhao, P., & Hu, T. (2025). Ebicop: Ensemble bivariate copulas for modeling multivariate cyber data breach risks.The Annals of Applied Statistics, 19(1), 566–585

2025
[19]

(2024).A framework for digital asset risks with insurance applications

Li, Z., Su, J., Xu, M., & Yuen, J. (2024).A framework for digital asset risks with insurance applications. arXiv preprint arXiv:2408.17227

2024 arXiv
[20]

Liu, Z., Wei, W., & Wang, L. (2021). An extreme value theory-based catastrophe bond design for cyber risk management of power systems.IEEE Transactions on Smart Grid, 13(2), 1516–1528. 42

2021
[21]

J., Frey, R., & Embrechts, P

McNeil, A. J., Frey, R., & Embrechts, P. (2015).Quantitative Risk Management: Concepts, Techniques and Tools-revised edition. Princeton university press

2015
[22]

& Vatter, T

Nagler, T. & Vatter, T. (2022). rvinecopulib: High performance algorithms for vine copula modeling.R package version

2022
[23]

(2024).Beazley$300m cyber cat bond via polestar re.https://www.artemis.bm/ news/

News, A. (2024).Beazley$300m cyber cat bond via polestar re.https://www.artemis.bm/ news/. [Accessed 10 September 2025]

2024
[24]

K., Joe, H., & Li, H

Nikoloulopoulos, A. K., Joe, H., & Li, H. (2012). Vine copulas with asymmetric tail dependence and applications to financial return data.Computational Statistics & Data Analysis, 56(11), 3659–3673

2012
[25]

(2024).Catalysing cyber risk transfer to capital markets: Catastrophe bonds and be- yond.https://www.genevaassociation.org/sites/default/files/2024-12/cyber_ils_ report_1213.pdf

Pain, D. (2024).Catalysing cyber risk transfer to capital markets: Catastrophe bonds and be- yond.https://www.genevaassociation.org/sites/default/files/2024-12/cyber_ils_ report_1213.pdf. [Accessed 15 September 2025]

2024
[26]

Peng, C., Xu, M., Xu, S., & Hu, T. (2018). Modeling multivariate cybersecurity risks.Journal of Applied Statistics, 45(15), 2718–2740

2018
[27]

(1987).Real and complex analysis

Rudin, W. (1987).Real and complex analysis. McGraw-Hill, Inc

1987
[28]

Shao, J., Pantelous, A., & Papaioannou, A. D. (2015). Catastrophe risk bonds with applications to earthquakes.European Actuarial Journal, 5(1), 113–138

2015
[29]

& Yang, L

Shi, P. & Yang, L. (2018). Pair copula constructions for insurance experience rating.Journal of the American Statistical Association, 113(521), 122–133

2018
[30]

Tsay, R. S. (2010).Analysis of financial time series. John Wiley & Sons

2010
[31]

& Zhang, Y

Xu, M. & Zhang, Y. (2021). Data breach cat bonds: Modeling and pricing.North American Actuarial Journal, 25(4), 543–561

2021
[32]

Zhou, L., Xiong, X., Ernstberger, J., Chaliasos, S., Wang, Z., Wang, Y., Qin, K., Wattenhofer, R., Song, D., & Gervais, A. (2023). Sok: Decentralized finance (defi) attacks.2023 IEEE Symposium on Security and Privacy (SP), 2444–2461. 43

2023

Pith tools

Reviewed July 9, 2026 · model on record in the stance chip above.

[1] [1]

Barrieu, P., Braun, A., & Makariou, D. (2024). Catastrophe bonds.Handbook of Insurance: Volume I: Third Edition, 169–195. Springer

work page 2024

[2] [2]

W., Gaur, V., & Giesecke, K

Biais, B., Capponi, A., Cong, L. W., Gaur, V., & Giesecke, K. (2023). Advances in blockchain and crypto economics.Management Science, 69(11), 6417–6426

work page 2023

[3] [3]

(2013).Convergence of probability measures

Billingsley, P. (2013).Convergence of probability measures. John Wiley & Sons. Bj¨ ork, T. (2009).Arbitrage theory in continuous time. Oxford university press

work page 2013

[4] [4]

Braun, A., Eling, M., & Jaenicke, C. (2023). Cyber insurance-linked securities.ASTIN Bulletin: The Journal of the IAA, 53(3), 684–705

work page 2023

[5] [5]

(2020).Insurance-linked securities primer.https://content.naic.org/sites/ default/files/capital-markets-primer-linked-securities.pdf

Carelus, J.-B. (2020).Insurance-linked securities primer.https://content.naic.org/sites/ default/files/capital-markets-primer-linked-securities.pdf. [Accessed 10 Septem- ber 2025]. F¨ ollmer, H. & Schweizer, M. (2010). Minimal martingale measure.Encyclopedia of Quantitative Finance, 3, 1200–1204. F¨ ollmer, H., Schweizer, M., et al. (1990).Hedging of cont...

work page 2020

[6] [6]

Gan, J., Tsoukalas, G., & Netessine, S. (2023). Decentralized platforms: Governance, toke- nomics, and ico design.Management Science, 69(11), 6667–6683

work page 2023

[7] [7]

Ghalanos, A. (2020). Introduction to the rugarch package.R vignette

work page 2020

[8] [8]

& Ferreira, A

Haan, L. & Ferreira, A. (2006).Extreme value theory: an introduction. Springer. 41 HKSAR (2021).Estimates of expenditure 2021–22: Financial services and the treasury bureau. https://www.legco.gov.hk/yr20-21/english/fc/fc/w_q/fstb-fs-e.pdf. [Accessed 10 September 2025]

work page 2006