pith. sign in

arxiv: 1907.03044 · v1 · pith:Y57ZPVHKnew · submitted 2019-07-05 · 🪐 quant-ph

Credit Risk Analysis using Quantum Computers

Daniel J. Egger , Ricardo Gac\'ia Guti\'errez , Jordi Cahu\'e Mestre , Stefan Woerner This is my paper

Pith reviewed 2026-05-25 01:53 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum algorithmcredit riskeconomic capitalvalue at riskmonte carloquantum amplitude estimationfault-tolerant quantum computing
0
0 comments X

The pith

A quantum algorithm estimates the economic capital requirement for a loss distribution more efficiently than classical Monte Carlo simulations.

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

The paper develops a quantum method to calculate how much capital a financial institution needs to hold against potential credit losses at a high confidence level. This metric, the difference between value at risk and expected loss, is key to assessing solvency. By loading the loss distribution into a quantum state and using quantum amplitude estimation, the approach aims to reduce the number of samples needed compared to running many simulations classically. The authors test it on a realistic portfolio and project the hardware resources required for larger instances.

Core claim

We present a quantum algorithm that computes the economic capital requirement by estimating the value at risk and expected loss of a credit portfolio loss distribution prepared as a quantum state, achieving better scaling than classical Monte Carlo methods, with explicit resource estimates for qubits and circuit depth on future fault-tolerant hardware.

What carries the argument

The quantum algorithm that prepares the loss distribution as a quantum state and applies amplitude estimation to compute the tail probabilities defining the economic capital.

Load-bearing premise

The loss distribution of the credit portfolio can be prepared efficiently as a quantum state without requiring resources that negate the quantum advantage.

What would settle it

Running the quantum circuit on a fault-tolerant device and measuring whether the estimated capital requirement matches the classical result within the predicted runtime and error bounds.

Figures

Figures reproduced from arXiv: 1907.03044 by Daniel J. Egger, Jordi Cahu\'e Mestre, Ricardo Gac\'ia Guti\'errez, Stefan Woerner.

Figure 1
Figure 1. Figure 1: FIG. 1. High level circuit of the operator [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: , scales as O(nzK), i.e. linear in the number of as￾sets. By adding O(K) ancilla qubits, the scaling of the circuit depth can be reduced to O(log K), which can lead to a potential speed-up. The additional qubits provide the compute space to perform more operations in par￾allel. Depending on the number of available qubits and the complexity of the rest of the algorithm, the number of ancillas can also be se… view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Cumulative distribution function (left) of total loss [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. The quantum circuit of amplitude estimation. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
read the original abstract

We present and analyze a quantum algorithm to estimate credit risk more efficiently than Monte Carlo simulations can do on classical computers. More precisely, we estimate the economic capital requirement, i.e. the difference between the Value at Risk and the expected value of a given loss distribution. The economic capital requirement is an important risk metric because it summarizes the amount of capital required to remain solvent at a given confidence level. We implement this problem for a realistic loss distribution and analyze its scaling to a realistic problem size. In particular, we provide estimates of the total number of required qubits, the expected circuit depth, and how this translates into an expected runtime under reasonable assumptions on future fault-tolerant quantum hardware.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript presents and analyzes a quantum algorithm, based on amplitude estimation, to compute the economic capital requirement (Value-at-Risk minus expected loss) for a credit portfolio loss distribution. It implements the approach for a realistic multi-obligor loss model, derives estimates of total qubit count and circuit depth, and translates these into projected runtimes on future fault-tolerant quantum hardware under stated assumptions, claiming a quadratic improvement over classical Monte Carlo sampling.

Significance. If the state-preparation and hardware assumptions hold, the work supplies one of the first end-to-end resource analyses for a finance-relevant quantum routine at realistic scale, including explicit qubit and depth figures that allow direct comparison with classical methods. Such concrete estimates are a strength for assessing whether quadratic sampling speedups can become practical.

major comments (2)
  1. [§4] §4 (Loss distribution encoding and state preparation): the circuit depth and qubit overhead for preparing the quantum state that encodes the loss distribution of a multi-obligor portfolio are not bounded as a function of the number of assets or risk factors; because this cost is added to every amplitude-estimation query, the claimed overall quadratic advantage over Monte Carlo cannot be verified without an explicit scaling argument.
  2. [§6] §6 (Runtime translation): the mapping from circuit depth to wall-clock time invokes specific gate durations (e.g., 10 ns) and error-correction overhead factors without a sensitivity analysis; if realistic overheads are larger, the projected runtime advantage disappears, rendering the practical-advantage claim dependent on unverified hardware parameters.
minor comments (2)
  1. [§2] Notation for the loss random variable L and the quantile level α is introduced without a dedicated table or equation reference, making cross-references in later sections harder to follow.
  2. [Figure 3] Figure 3 (circuit diagram) lacks a caption explaining the register partitioning (loss register vs. ancillary qubits), which obscures the resource count.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and positive assessment of the significance of our work. We address the two major comments below.

read point-by-point responses

  1. Referee: [§4] §4 (Loss distribution encoding and state preparation): the circuit depth and qubit overhead for preparing the quantum state that encodes the loss distribution of a multi-obligor portfolio are not bounded as a function of the number of assets or risk factors; because this cost is added to every amplitude-estimation query, the claimed overall quadratic advantage over Monte Carlo cannot be verified without an explicit scaling argument.

    Authors: We agree that providing an explicit scaling bound for the state preparation circuit would strengthen the verification of the quadratic advantage. The manuscript provides detailed qubit and depth estimates for a concrete multi-obligor portfolio in Section 4, based on the chosen loss model. However, to fully address this point, we will revise the manuscript to include an analysis of how the state preparation scales with the number of assets and risk factors, demonstrating that the preparation cost is polynomial and does not negate the quadratic speedup from amplitude estimation. revision: yes

  2. Referee: [§6] §6 (Runtime translation): the mapping from circuit depth to wall-clock time invokes specific gate durations (e.g., 10 ns) and error-correction overhead factors without a sensitivity analysis; if realistic overheads are larger, the projected runtime advantage disappears, rendering the practical-advantage claim dependent on unverified hardware parameters.

    Authors: We acknowledge the validity of this observation. The runtime estimates in Section 6 are based on stated assumptions about future hardware. In the revised manuscript, we will add a sensitivity analysis varying the key parameters such as gate durations and overhead factors to show the range of conditions under which the advantage persists. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation applies standard amplitude estimation to external loss metric

full rationale

The paper's core claim is that quantum amplitude estimation can compute economic capital (VaR minus expectation) from a loss distribution more efficiently than classical Monte Carlo. This rests on two external premises—the efficient preparation of the loss distribution as a quantum state and optimistic fault-tolerant hardware parameters—neither of which is derived from or defined in terms of the algorithm's output. No equations or steps in the abstract reduce a prediction to a fitted input, rename a known result, or rely on a self-citation chain for uniqueness. The approach is therefore self-contained against independent benchmarks (classical Monte Carlo and prior amplitude estimation results).

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; full text would be required to audit the state-preparation subroutine and hardware assumptions.

pith-pipeline@v0.9.0 · 5648 in / 1027 out tokens · 15257 ms · 2026-05-25T01:53:11.446228+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. On the complexity of quantum numerical integration: an angle-structure characterization

    quant-ph 2026-04 unverdicted novelty 7.0

    Low-degree multilinear angle maps enable O(ε^{-1} log(1/ε)) quantum gate complexity for numerical integration on [0,1], with unconditional separations from classical quadrature for certain low-regularity functions.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    Credit Risk Analysis using Quantum Computers

    or to price financial derivatives [15] with a quadratic speedup. In Section II, we formally define the economic capi- tal requirement as well as the two different uncertainty models considered. In Section III, we build on previous work [11] and discuss how to implement the quantum ∗ wor@zurich.ibm.com algorithms on a gate based quantum computer. In Sec- tion...

    work page internal anchor Pith review Pith/arXiv arXiv 1907

  2. [2]

    To achieve the loga- rithmic scaling, a linear number of ancilla qubits needs to be added

    where a construction of adders and comparators is introduced and analyzed in detail. To achieve the loga- rithmic scaling, a linear number of ancilla qubits needs to be added

    work page

  3. [3]

    An approach to economic capital for financial services firms,

    Bruce Porteous, Louise McCulloch, and PradipTapadar, “An approach to economic capital for financial services firms,” Risk , 28–31 (2003)

    work page 2003

  4. [4]

    Managing post-convergence risks in fi- nancial conglomerates,

    Bruce Porteous, “Managing post-convergence risks in fi- nancial conglomerates,” Risk , 21–24 (2002)

    work page 2002

  5. [5]

    Principles for the management of credit risk,

    Risk Management Group of the Basel Committee on Banking Supervision, “Principles for the management of credit risk,” (2000)

    work page 2000

  6. [6]

    Sylvain Bouteillé and Diane Coogan-Pushner,The Hand- book of Credit Risk Management(2013) p. 322

    work page 2013

  7. [7]

    Paul Glasserman,Monte Carlo Methods in Financial En- gineering (Springer-Verlag New York, 2003) p. 596

    work page 2003

  8. [8]

    Importance sampling for portfolio credit risk,

    Paul Glasserman and Jingyi Li, “Importance sampling for portfolio credit risk,” Manag. Sci.51, 11 (2005)

    work page 2005

  9. [9]

    Nielsen and Isaac L

    Michael A. Nielsen and Isaac L. Chuang,Cambridge Uni- versity Press(2010) p. 702

    work page 2010

  10. [10]

    Quantum optimization using variational algorithms on near-term quantum devices,

    N. Moll, P. Barkoutsos, L. S. Bishop, J. M. Chow, A. Cross, D. J. Egger, S. Filipp, A. Fuhrer, J. M. Gambetta, M. Ganzhorn, A. Kandala, A. Mezzacapo, P. Müller, W. Riess, G. Salis, J. Smolin, I. Tavernelli, and K. Temme, “Quantum optimization using variational algorithms on near-term quantum devices,” Quantum Science and Technology3, 030503 (2018)

    work page 2018

  11. [11]

    Error mitigation extends the computational reach of a noisy quantum processor,

    Abhinav Kandala, Kristan Temme, Antonio D. Corcoles, Antonio Mezzacapo, Jerry M. Chow, and Jay M. Gam- betta, “Error mitigation extends the computational reach of a noisy quantum processor,” Nature 567, 491–495 (2018)

    work page 2018

  12. [12]

    Supervised learning with quantum-enhanced feature spaces,

    Vojtech Havlicek, Antonio D. Corcoles, Kristan Temme, Aram W. Harrow, Abhinav Kandala, Jerry M. Chow, and Jay M. Gambetta, “Supervised learning with quantum-enhanced feature spaces,” Nature 567, 209 – 212 (2019)

    work page 2019

  13. [13]

    Quantumriskanal- ysis,

    StefanWoernerandDanielJ.Egger,“Quantumriskanal- ysis,” npj Quantum Information5, 15 (2019)

    work page 2019

  14. [14]

    Quantum computational finance: Monte carlo pricing of financial derivatives,

    Patrick Rebentrost, Brajesh Gupt, and Thomas R. Bromley, “Quantum computational finance: Monte carlo pricing of financial derivatives,” Phys. Rev. A98, 022321 (2018)

    work page 2018

  15. [15]

    Towards pricing financial derivatives with an ibm quantum com- puter,

    Ana Martin, Bruno Candelas, Angel Rodriguez-Rozas, Jose D. Martin-Guerrero, Xi Chen, Lucas Lamata, Ro- man Orus, Enrique Solano, and Mikel Sanz, “Towards pricing financial derivatives with an ibm quantum com- puter,” arXiv:1904.05803

    work page arXiv 1904

  16. [16]

    Quantum computing for finance: Overview and prospects,

    Roman Orus, Samuel Mugel, and Enrique Lizaso, “Quantum computing for finance: Overview and prospects,” Reviews in Physics4, 100028 (2019)

    work page 2019

  17. [17]

    Option Pricing using Quantum Computers,

    Nikitas Stamatopoulos, Daniel J. Egger, Yue Sun, Christa Zoufal, Raban Iten, Ning Shen, and Stefan Woerner, “Option Pricing using Quantum Computers,” arXiv:1905.02666

    work page arXiv 1905

  18. [18]

    Basel III: A global regulatory framework for more resilient banks and banking systems,

    Basel Committee on Banking Supervision, “Basel III: A global regulatory framework for more resilient banks and banking systems,” (2010)

    work page 2010

  19. [19]

    Regulatory capital modelling for credit risk,

    Marek Rutkowski and Silvio Tarca, “Regulatory capital modelling for credit risk,” International Journal of The- oretical and Applied Finance18, 1550034 (2015)

    work page 2015

  20. [20]

    International convergence of capital measurement and capital stan- dards,

    Basel Committee on Banking Supervision, “International convergence of capital measurement and capital stan- dards,” (2006)

    work page 2006

  21. [21]

    Revisions to the Basel II market risk framework,

    Basel Committee on Banking Supervision, “Revisions to the Basel II market risk framework,” (2009)

    work page 2009

  22. [22]

    Quantum Amplitude Amplification and Estima- tion,

    Gilles Brassard, Peter Hoyer, Michele Mosca, and Alain Tapp, “Quantum Amplitude Amplification and Estima- tion,” Contemporary Mathematics305, 53–74 (2002)

    work page 2002

  23. [23]

    Creating superpositions that correspond to efficiently integrable probability dis- tributions,

    Lov Grover and Terry Rudolph, “Creating superpositions that correspond to efficiently integrable probability dis- tributions,” (2002), arXiv:0208112

    work page 2002

  24. [24]

    Qiskit: An open-source framework for quantum computing,

    Gadi Aleksandrowicz, Thomas Alexander, Panagi- otis Barkoutsos, Luciano Bello, Yael Ben-Haim, David Bucher, Francisco Jose Cabrera-Hernádez, Jorge Carballo-Franquis, Adrian Chen, Chun-Fu Chen, Jerry M. Chow, Antonio D. Córcoles-Gonzales, Abigail J. Cross, Andrew Cross, Juan Cruz-Benito, Chris Culver, Salvador De La Puente González, Enrique De La Torre, De...

    work page 2019

  25. [25]

    Fault-tolerant quantum computation,

    Peter W. Shor, “Fault-tolerant quantum computation,” inProceedings of the 37th Annual Symposium on Founda- tions of Computer Science, FOCS ’96 (IEEE Computer Society, Washington, DC, USA, 1996) pp. 56–

    work page 1996

  26. [26]

    Fault-tolerant quantum computation by anyons,

    A.Yu. Kitaev, “Fault-tolerant quantum computation by anyons,” Annals of Physics303, 2 – 30 (2003)

    work page 2003

  27. [27]

    Surface codes: Towards practi- cal large-scale quantum computation,

    Austin G. Fowler, Matteo Mariantoni, John M. Martinis, and Andrew N. Cleland, “Surface codes: Towards practi- cal large-scale quantum computation,” Phys. Rev. A86, 032324 (2012)

    work page 2012

  28. [28]

    Magic-state distilla- tion with low overhead,

    Sergey Bravyi and Jeongwan Haah, “Magic-state distilla- tion with low overhead,” Phys. Rev. A86, 052329 (2012)

    work page 2012

  29. [29]

    Quantum circuits of t-depth one,

    Peter Selinger, “Quantum circuits of t-depth one,” Phys. Rev. A 87, 042302 (2013)

    work page 2013

  30. [30]

    Low overhead quantum computation using lattice surgery,

    Austin G. Fowler and Craig Gidney, “Low overhead quantum computation using lattice surgery,” (2018), arXiv:1808.06709

    work page arXiv 2018

  31. [31]

    Approxi- mate quantum fourier transform with o(nlog(n)) t gates,

    Yunseong Nam, Yuan Su, and Dmitri Maslov, “Approxi- mate quantum fourier transform with o(nlog(n)) t gates,” (2018), arXiv:1803.04933

    work page arXiv 2018

  32. [32]

    Am- plitude estimation without phase estimation,

    Yohichi Suzuki, Shumpei Uno, Rudy Raymond, Tomoki Tanaka, Tamiya Onodera, and Naoki Yamamoto, “Am- plitude estimation without phase estimation,” (2019), arXiv:1904.10246

    work page arXiv 2019

  33. [33]

    On the advantages of using relative phase toffolis with an application to multiple control tof- foli optimization,

    Dmitri Maslov, “On the advantages of using relative phase toffolis with an application to multiple control tof- foli optimization,” Phys. Rev. A93, 022311 (2016)

    work page 2016

  34. [34]

    Lin- ear and logarithmic time compositions of quantum many- body operators,

    F. Motzoi, M. P. Kaicher, and F. K. Wilhelm, “Lin- ear and logarithmic time compositions of quantum many- body operators,” Phys. Rev. Lett.119, 160503 (2017)

    work page 2017

  35. [35]

    A confi- dence interval procedure for expected shortfall risk mea- surement via two-level simulation,

    Hai Lan, Barry L. Nelson, and Jeremy Staum, “A confi- dence interval procedure for expected shortfall risk mea- surement via two-level simulation,” Operations Research 58, 1481–1490 (2010)

    work page 2010

  36. [36]

    Nested mc-based risk measure- ment of complex portfolios: Acceleration and energy ef- ficiency,

    Sascha Desmettre, Ralf Korn, Javier Alejandro Varela, and Norbert Wehn, “Nested mc-based risk measure- ment of complex portfolios: Acceleration and energy ef- ficiency,” Risks 4, 36 (2016)

    work page 2016

  37. [37]

    Large-scale data-driven financial risk modeling using big data technology,

    Kurt Stockinger, Jonas Heitz, Nils Andri Bundi, and Wolfgang Breymann, “Large-scale data-driven financial risk modeling using big data technology,” International conference on Big Data computing, applications and technologies (2018)

    work page 2018

  38. [38]

    Fast quantum algorithms for numerical integrals and stochastic pro- cesses,

    Daniel S Abrams and Colin P Williams, “Fast quantum algorithms for numerical integrals and stochastic pro- cesses,” (1999), arXiv:9908083

    work page 1999

  39. [39]

    Quantum speedup of monte carlo methods,

    Ashley Montanaro, “Quantum speedup of monte carlo methods,” Proceedings of the Royal Society A: Math- ematical, Physical and Engineering Sciences 471, 20150301 (2015)

    work page 2015

  40. [40]

    Quantum measurements and the Abelian Stabilizer Problem,

    A. Yu. Kitaev, “Quantum measurements and the Abelian Stabilizer Problem,” (1995), arXiv:9511026

    work page 1995

  41. [41]

    Efficient Monte Carlo Methods for Value-at- Risk,

    Paul Glasserman, Philip Heidelberger, and Perwez Sha- habuddin, “Efficient Monte Carlo Methods for Value-at- Risk,” inMastering Risk, Vol. 2 (2000) pp. 5–18

    work page 2000

  42. [42]

    A meet-in-the-middle algorithm for fast synthesis of depth-optimal quantum circuits,

    Matthew Amy, Dmitri Maslov, Michele Mosca, and Martin Roetteler, “A meet-in-the-middle algorithm for fast synthesis of depth-optimal quantum circuits,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems32, 818–830 (2013)

    work page 2013

  43. [43]

    Practical approximation of single-qubit unitaries by single-qubit quantum Clifford and T circuits,

    Vadym Kliuchnikov, Dmitri Maslov, and Michele Mosca, “Practical approximation of single-qubit unitaries by single-qubit quantum Clifford and T circuits,” IEEE Transactions on Computers65, 161–172 (2012)

    work page 2012

  44. [44]

    A new quantum ripple-carry addition circuit,

    Steven A Cuccaro, Thomas G Draper, Samuel A Kutin, and David Petrie Moulton, “A new quantum ripple-carry addition circuit,” (2004), arXiv:0410184

    work page 2004

  45. [45]

    A logarithmic-depth quantum carry- lookahead adder,

    ThomasG.Draper, SamuelA.Kutin, EricM.Rains, and Krysta M. Svore, “A logarithmic-depth quantum carry- lookahead adder,” Quantum Information and Computa- tion 6, 351–369 (2006)

    work page 2006