pith. sign in

arxiv: 2606.24309 · v1 · pith:JL2NCPXHnew · submitted 2026-06-23 · 💱 q-fin.CP

Randomized Neural Networks for estimation of exposure profiles and Credit Valuation Adjustment (CVA) for American Equity Options

Isidro Moroso Varona , Jakub Micha\'nk\'ow , Pawe{\l} Sakowski This is my paper

Pith reviewed 2026-06-25 21:54 UTC · model grok-4.3

classification 💱 q-fin.CP
keywords randomized neural networksAmerican optionsexposure profilesCVAMonte Carlo simulationleast-squares Monte Carlohigh-dimensional problems
0
0 comments X

The pith

Randomized neural networks estimate exposure and CVA for American options with the same accuracy as least-squares Monte Carlo but lower cost in high dimensions.

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

The paper tests randomized feedforward neural networks inside a Monte Carlo simulation to compute exposure profiles and unilateral CVA for American equity options. Experiments cover both Black-Scholes and Heston dynamics, single-asset and multi-asset portfolios with netting, and compare results directly to the least-squares Monte Carlo benchmark. The neural-network estimates converge to the benchmark values for prices, expected exposure, potential future exposure, and CVA. The computational advantage grows with the number of assets because the randomized network avoids the growing regression burden of least-squares Monte Carlo.

Core claim

The randomized feedforward neural network approach preserves convergence to the LSM benchmark when it is extended from pricing to exposure and CVA estimation, while its main advantage appears in high-dimensional problems, where it scales more efficiently and leads to lower computational cost.

What carries the argument

Randomized feedforward neural network trained on simulated paths to estimate continuation values for American exercise decisions inside Monte Carlo exposure and CVA calculations.

If this is right

  • The method produces usable exposure and CVA numbers for both Black-Scholes and Heston American options.
  • Portfolio netting effects are captured at the same accuracy level as LSM.
  • Pathwise convergence and sensitivity tests remain stable when the network replaces LSM regression.
  • Computational cost stays lower than LSM once the dimension exceeds a few assets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same network could replace LSM regression steps in other path-dependent claims such as Bermudan swaptions.
  • Real-time risk systems for large equity baskets might become feasible if training cost stays sub-linear in dimension.
  • Extending the approach to stochastic interest rates would require only adding the rate factors to the network inputs.

Load-bearing premise

Once trained on the same paths used by LSM, the randomized network produces continuation-value estimates whose statistical properties stay close enough to LSM that downstream exposure and CVA numbers do not diverge materially.

What would settle it

Run the identical high-dimensional test set and observe whether the neural-network CVA or exposure profile deviates from the LSM benchmark by more than Monte Carlo sampling error.

Figures

Figures reproduced from arXiv: 2606.24309 by Isidro Moroso Varona, Jakub Micha\'nk\'ow, Pawe{\l} Sakowski.

Figure 1
Figure 1. Figure 1: Expected exposure curves for the vanilla American call under Black-Scholes dynamics. [PITH_FULL_IMAGE:figures/full_fig_p040_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Expected exposure curves for the vanilla American call under Heston dynamics. [PITH_FULL_IMAGE:figures/full_fig_p041_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: EE and PFE comparison for the vanilla American call. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p042_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: EE curves for American max-call options under Black-Scholes dynamics across dimen [PITH_FULL_IMAGE:figures/full_fig_p044_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: EE curves for American max-call options under Heston dynamics across dimensions. [PITH_FULL_IMAGE:figures/full_fig_p045_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: EE comparison for the 50-dimensional Heston max-call. LSM with 10.000 paths, LSM [PITH_FULL_IMAGE:figures/full_fig_p046_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: PFE curve for American max-call options under Black-Scholes dynamics across di [PITH_FULL_IMAGE:figures/full_fig_p047_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Price convergence with respect to the number of paths for the 5-dimensional Heston [PITH_FULL_IMAGE:figures/full_fig_p048_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Price convergence with respect to the number of paths for the 10-dimensional Heston [PITH_FULL_IMAGE:figures/full_fig_p048_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Price convergence with respect to the number of paths for the 25-dimensional Heston [PITH_FULL_IMAGE:figures/full_fig_p049_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Hidden-size sensitivity for the Black-Scholes model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p049_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Hidden-size sensitivity for the Heston model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p050_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Regularization sensitivity for the Black-Scholes model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p050_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Regularization sensitivity for the Heston model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p051_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Scaling factor sensitivity for the Black-Scholes model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p051_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Scaling factor sensitivity for the Heston model. Source: Own elaboration. [PITH_FULL_IMAGE:figures/full_fig_p052_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Trade-level option values profiles for the five-trade Heston netting set. Source: Own [PITH_FULL_IMAGE:figures/full_fig_p054_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: Portfolio expected exposure under shared 50-dimensional Heston paths. Source: Own [PITH_FULL_IMAGE:figures/full_fig_p056_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Incremental and cumulative portfolio CVA in the base credit scenario. Source: Own [PITH_FULL_IMAGE:figures/full_fig_p057_19.png] view at source ↗
read the original abstract

This thesis studies the use of randomized neural networks for the estimation of exposure profiles and unilateral CVA of American options within a Monte Carlo framework. The analysis is carried out separately under both Black-Scholes and Heston dynamics, combining American option valuation, expected exposure and potential future exposure estimation, and unilateral CVA calculation with portfolio netting effects. The numerical experiment compares this approach with the classical Least-Squares Monte Carlo (LSM) used as a benchmark in both low-dimensional single-asset and high-dimensional multi-asset scenarios, and also includes a path convergence test and a sensitivity analysis. The results show that the randomized feedforward neural network approach preserves convergence to the LSM benchmark when it is extended from pricing to exposure and CVA estimation, while its main advantage appears in high-dimensional problems, where it scales more efficiently and leads to lower computational cost. These results support the use of randomized neural networks as a useful alternative for exposure and CVA estimation in high-dimensional American-style options.

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 / 3 minor

Summary. The manuscript proposes the use of randomized feedforward neural networks within a Monte Carlo simulation framework to estimate exposure profiles (EE, PFE) and unilateral CVA for American equity options. The method is tested separately under Black-Scholes and Heston dynamics, for single-asset and multi-asset portfolios (with netting), and is benchmarked against Least-Squares Monte Carlo (LSM). Numerical results include path-convergence tests and sensitivity analysis; the central claim is that the randomized NN approach converges to the LSM benchmark while offering computational scaling advantages in high-dimensional settings.

Significance. If the empirical convergence holds, the work supplies a practical, lower-cost alternative to LSM for CVA and exposure calculations on American-style claims in high-dimensional equity portfolios. The explicit inclusion of both dynamics, netting, and path-convergence diagnostics strengthens the applicability claim for risk-management use cases.

major comments (2)
  1. [§4.2] §4.2 (high-dimensional experiments): the reported CPU-time savings versus LSM are stated without accompanying standard errors or number of independent runs; it is therefore unclear whether the observed speed-up is statistically distinguishable from run-to-run Monte Carlo variability.
  2. [Table 5] Table 5 (CVA values under Heston with netting): the absolute differences between NN and LSM are on the order of 1–3 % of the LSM value, yet no formal statistical test (e.g., paired t-test across paths) is supplied to confirm that these differences are consistent with Monte Carlo noise rather than systematic bias in the continuation-value estimator.
minor comments (3)
  1. [Abstract] The abstract states that “a path-convergence test … were performed” but does not report the observed convergence rates or the number of paths used; adding a short quantitative summary would improve readability.
  2. [§3.1] Notation for the randomized NN (weights drawn from which distribution, activation function, number of hidden units) is introduced only in §3.1; a compact table summarizing the hyper-parameters used in each experiment would aid reproducibility.
  3. [Figure 3] Figure 3 (exposure profiles) uses different y-axis scales across panels; aligning the scales or adding a common reference line would make visual comparison with LSM easier.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address the two major comments point by point below. We agree that the suggested additions of statistical information will strengthen the manuscript and plan to incorporate them.

read point-by-point responses

  1. Referee: [§4.2] §4.2 (high-dimensional experiments): the reported CPU-time savings versus LSM are stated without accompanying standard errors or number of independent runs; it is therefore unclear whether the observed speed-up is statistically distinguishable from run-to-run Monte Carlo variability.

    Authors: We agree that reporting the number of independent runs and standard errors is necessary for a rigorous interpretation of the timing results. In the revised manuscript we will state the number of independent timing runs performed in §4.2 and include standard errors for the reported CPU times, allowing readers to confirm that the observed speed-ups exceed run-to-run Monte Carlo variability. revision: yes

  2. Referee: [Table 5] Table 5 (CVA values under Heston with netting): the absolute differences between NN and LSM are on the order of 1–3 % of the LSM value, yet no formal statistical test (e.g., paired t-test across paths) is supplied to confirm that these differences are consistent with Monte Carlo noise rather than systematic bias in the continuation-value estimator.

    Authors: We accept that a formal test would provide additional reassurance. In the revision we will compute and report a paired t-test (or equivalent) on the CVA estimates obtained from multiple independent Monte Carlo path sets, demonstrating that the observed 1–3 % differences lie within the range attributable to Monte Carlo noise rather than estimator bias. The test results and updated discussion will be added to the presentation of Table 5. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical results rest on external LSM benchmark

full rationale

The paper's core claims are established through direct numerical comparison of randomized NN continuation values against the independent LSM benchmark on the same simulated paths, under both Black-Scholes and Heston dynamics, including path-convergence tests and sensitivity analysis for exposure and CVA. No derivation step reduces a reported CVA or exposure statistic to a fitted parameter or self-defined quantity by construction; the randomized NN is trained once and then evaluated for statistical closeness to LSM, which is an external method. No load-bearing self-citation chains or uniqueness theorems imported from the authors' prior work appear in the derivation. The work is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no explicit free parameters, axioms, or invented entities. The method implicitly assumes that randomized networks can be trained to approximate continuation values without introducing bias that propagates into exposure statistics.

pith-pipeline@v0.9.1-grok · 5718 in / 1107 out tokens · 22247 ms · 2026-06-25T21:54:14.825512+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

28 extracted references

  1. [1]

    and Broadie, M

    Andersen, L. and Broadie, M. (2004). A primal-dual simulation algorithm for pricing multidi- mensional american options.Management Science, 50(9):1222–1234

    2004

  2. [2]

    and Oosterlee, C

    Andersson, K. and Oosterlee, C. W. (2021). A deep learning approach for computations of expo- sure profiles for high-dimensional bermudan options.Applied Mathematics and Computation, 408:126332. Bank for International Settlements (2018). Counterparty credit risk in basel iii – executive summary. FSI Executive Summary. Accessed 2026-04-15. Bank for Interna...

    2021

  3. [3]

    Becker, S., Cheridito, P., and Jentzen, A. (2019). Deep optimal stopping.Journal of Machine Learning Research, 20(74):1–25

    2019

  4. [4]

    and Scholes, M

    Black, F. and Scholes, M. (1973). The pricing of options and corporate liabilities.Journal of Political Economy, 81(3):637–654

    1973

  5. [5]

    Boyle, P. P. (1977). Options: A monte carlo approach.Journal of Financial Economics, 4(3):323–338

    1977

  6. [6]

    and Alfonsi, A

    Brigo, D. and Alfonsi, A. (2005). Credit default swap calibration and derivatives pricing with the ssrd stochastic intensity model.Finance and Stochastics, 9(1):29–42

    2005

  7. [7]

    (2013).Counterparty Credit Risk, Collateral and Funding: With Pricing Cases for All Asset Classes

    Brigo, D., Morini, M., and Pallavicini, A. (2013).Counterparty Credit Risk, Collateral and Funding: With Pricing Cases for All Asset Classes. John Wiley & Sons, Chichester

    2013

  8. [8]

    and Cao, M

    Broadie, M. and Cao, M. (2008). Improved lower and upper bound algorithms for pricing american options by simulation.Quantitative Finance, 8(8):845–861

    2008

  9. [9]

    and Glasserman, P

    Broadie, M. and Glasserman, P. (2004). A stochastic mesh method for pricing high-dimensional american options.Journal of Computational Finance, 7(4):35–72

    2004

  10. [10]

    Cao, W., Wang, X., Ming, Z., and Gao, J. (2018). A review on neural networks with random weights.Neurocomputing, 275:278–287. Carri` ere, J. F. (1996). Valuation of the early-exercise price for options using simulations and nonparametric regression.Insurance: Mathematics and Economics, 19(1):19–30

    2018

  11. [11]

    (2009).Modelling, Pricing, and Hedging Counterparty Credit Exposure: A Technical Guide

    Cesari, G., Aquilina, J., Charpillon, N., Filipovic, Z., Lee, G., and Manda, I. (2009).Modelling, Pricing, and Hedging Counterparty Credit Exposure: A Technical Guide. Springer, Berlin and Heidelberg

    2009

  12. [12]

    Cortis, D. (2019). Evaluating credit counterparty risk of american options via monte carlo methods.Frontiers in Applied Mathematics and Statistics, 5:60

    2019

  13. [13]

    (2004).Monte Carlo Methods in Financial Engineering, volume 53 ofApplica- tions of Mathematics

    Glasserman, P. (2004).Monte Carlo Methods in Financial Engineering, volume 53 ofApplica- tions of Mathematics. Springer, New York

    2004

  14. [14]

    (2010).Counterparty Credit Risk: The New Challenge for Global Financial Markets

    Gregory, J. (2010).Counterparty Credit Risk: The New Challenge for Global Financial Markets. John Wiley & Sons, Chichester

    2010

  15. [15]

    (2015).The xVA Challenge: Counterparty Credit Risk, Funding, Collateral, and Capital

    Gregory, J. (2015).The xVA Challenge: Counterparty Credit Risk, Funding, Collateral, and Capital. John Wiley & Sons, Chichester. 59

    2015

  16. [16]

    Herrera, C., Krach, F., Ruyssen, P., and Teichmann, J. (2024). Optimal stopping via randomized neural networks.Frontiers of Mathematical Finance, 3(1):31–77

    2024

  17. [17]

    Heston, S. L. (1993). A closed-form solution for options with stochastic volatility with applica- tions to bond and currency options.The Review of Financial Studies, 6(2):327–343

    1993

  18. [18]

    Huang, G.-B., Chen, L., and Siew, C.-K. (2006). Universal approximation using incremental constructive feedforward networks with random hidden nodes.IEEE Transactions on Neural Networks, 17(4):879–892

    2006

  19. [19]

    Hull, J. C. (2012).Options, Futures, and Other Derivatives. Pearson, 8 edition

    2012

  20. [20]

    and Oosterlee, C

    Jain, S. and Oosterlee, C. W. (2015). The stochastic grid bundling method: Efficient pricing of bermudan options and their greeks.Applied Mathematics and Computation, 269:412–431

    2015

  21. [21]

    and Lelong, J

    Lapeyre, B. and Lelong, J. (2021). Neural network regression for bermudan option pricing. Monte Carlo Methods and Applications, 27(3):227–247

    2021

  22. [22]

    Longstaff, F. A. and Schwartz, E. S. (2001). Valuing american options by simulation: A simple least-squares approach.The Review of Financial Studies, 14(1):113–147. Lukoˇ seviˇ cius, M. and Jaeger, H. (2009). Reservoir computing approaches to recurrent neural network training.Computer Science Review, 3(3):127–149

    2001

  23. [23]

    Merton, R. C. (1973). Theory of rational option pricing.The Bell Journal of Economics and Management Science, 4(1):141–183

    1973

  24. [24]

    Rogers, L. C. G. (2002). Monte carlo valuation of american options.Mathematical Finance, 12(3):271–286

    2002

  25. [25]

    Schrauwen, B., Verstraeten, D., and Van Campenhout, J. (2007). An overview of reservoir computing: Theory, applications and implementations. InProceedings of the 15th European Symposium on Artificial Neural Networks, pages 471–482

    2007

  26. [26]

    Tsitsiklis, J. N. and Van Roy, B. (1999). Optimal stopping of markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives.IEEE Transactions on Automatic Control, 44(10):1840–1851

    1999

  27. [27]

    Tsitsiklis, J. N. and Van Roy, B. (2001). Regression methods for pricing complex american-style options.IEEE Transactions on Neural Networks, 12(4):694–703

    2001

  28. [28]

    and Pykhtin, M

    Zhu, S. and Pykhtin, M. (2007). A guide to modelling counterparty credit risk.GARP Risk Review, pages 16–22. Issue 37. 60 List of Tables 1 Vanilla American call results under Black-Scholes dynamics. . . . . . . . . . . . . 39 2 Vanilla American call results under Heston dynamics. . . . . . . . . . . . . . . . 40 3 American max-call results under Black-Sch...

    2007