pith. sign in

arxiv: 2604.08889 · v1 · submitted 2026-04-10 · 🧮 math.PR

Matrix Representations for Scale Functions of Spectrally Negative L\'evy Processes with Rational Jumps

Osvaldo Angtuncio Hern\'andez , Oscar Peralta This is my paper

Pith reviewed 2026-05-10 17:53 UTC · model grok-4.3

classification 🧮 math.PR
keywords scale functionsspectrally negative Lévy processesrational arrival processesmatrix representationsfluctuation theoryruin theorymatrix-exponential distributions
0
0 comments X

The pith

Spectrally negative Lévy processes with matrix-exponential jumps admit explicit scale function representations through rational arrival process embeddings.

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

The paper develops a method to obtain matrix representations for the scale functions of spectrally negative Lévy processes with rational jumps. It extends an earlier representation limited to phase-type jumps by embedding the original process into a stochastic fluid process modulated by a rational arrival process. Iterative schemes applied to this embedding produce a simple explicit formula for the q-scale function whose Laplace transform is the reciprocal of the Laplace exponent minus q. The construction recovers the known phase-type case as a special instance and applies directly to fluctuation problems such as exit times and ruin probabilities.

Core claim

By embedding the spectrally negative Lévy process into a stochastic fluid process modulated by a rational arrival process, the q-scale function receives an explicit matrix representation constructed via iterative schemes that generalize the phase-type formulas.

What carries the argument

The embedding of the Lévy process into a stochastic fluid process modulated by a rational arrival process, which converts the scale-function problem into matrix-analytic iteration.

If this is right

  • The scale function admits an explicit matrix representation for all matrix-exponential jump distributions.
  • Iterative schemes compute the scale function without requiring the phase-type restriction.
  • Exit probabilities and first-passage times inherit the same matrix form for this larger class of processes.
  • The representation reduces exactly to the earlier phase-type formula when the jumps are phase-type.

Where Pith is reading between the lines

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

  • The same embedding technique may produce matrix formulas for other fluctuation quantities such as resolvents or occupation measures.
  • Rational arrival process representations could link Lévy fluctuation theory to broader classes of Markov-modulated models used in queueing and risk analysis.
  • Numerical stability of the iterative schemes can be checked by comparing outputs against known closed-form scale functions for exponential or hyperexponential jumps.

Load-bearing premise

The scale-function identity for the original Lévy process remains unchanged after the embedding into the rational arrival process modulated fluid process.

What would settle it

For a concrete spectrally negative Lévy process whose jumps have a known matrix-exponential distribution, compute the Laplace transform of the formula produced by the iterative scheme and check whether it equals one divided by the Laplace exponent minus q.

Figures

Figures reproduced from arXiv: 2604.08889 by Oscar Peralta, Osvaldo Angtuncio Hern\'andez.

Figure 1
Figure 1. Figure 1: Path decomposition for hitting −x when σ = 0, d > 0. The process starts at zero, drifts upward (blue) with downward jumps (red). First downcrossing τ − {−x} has overshoot Z, followed by continuous upcrossing (green) to reach τ{−x} . By the spatial homogeneity of L´evy processes, the probability of reaching 0 from −Z is identical to reaching −Z from 0. Thus, evaluating the conditional law E−Z[1{τ{0} < eq}] … view at source ↗
Figure 2
Figure 2. Figure 2: Embedding of a spectrally negative L´evy process into a RAP-modulated fluid [PITH_FULL_IMAGE:figures/full_fig_p022_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Embedding of a spectrally negative L´evy process with Brownian component into [PITH_FULL_IMAGE:figures/full_fig_p024_3.png] view at source ↗
read the original abstract

For a spectrally negative L\'evy process with Laplace transform $\psi$, the $q$-scale function is characterized as the function whose Laplace transform is $(\psi(\cdot)-q)^{-1}$. It has applications in fluctuation theory, for example, exit problems and first hitting probabilities. It is also used in areas like ruin theory, risk theory, continuous state branching processes and optimal control. In this paper, we extend the scale function representation of Ivanovs (2021) from spectrally negative L\'evy processes with phase-type jumps to the general case of matrix-exponential jumps. The extension is non-trivial because the probabilistic arguments employed by Ivanovs rely on an embedding to a Markov-modulated Brownian motion, a framework that does not accommodate the algebraic generality of matrix-exponential distributions. We overcome this limitation by embedding the L\'evy process into a stochastic fluid process modulated by a rational arrival process (RAP), a class of continuous-valued Markov processes driven by orbit processes. This approach yields iterative schemes related to those of Ivanovs (2021) to provide a simple and explicit formula for the scale function. Our method gives the same fixed point when restricted to the phase-type case, and demonstrates the utility of orbit representations in analytical problems beyond the phase-type setting.

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

1 major / 0 minor

Summary. The manuscript extends the matrix representation of q-scale functions for spectrally negative Lévy processes from the phase-type jump case (Ivanovs 2021) to the more general matrix-exponential jump case. The extension is achieved by embedding the Lévy process into a stochastic fluid process modulated by a rational arrival process (RAP), leading to iterative schemes that provide an explicit formula for the scale function. The method is shown to recover the phase-type case as a fixed point.

Significance. If the central construction holds, the result offers a practical way to obtain explicit representations for scale functions beyond phase-type distributions, which are important in fluctuation theory, ruin theory, and related fields. The approach highlights the utility of RAP and orbit representations for analytical problems, generalizing previous probabilistic embeddings.

major comments (1)
  1. The claim that the scale function obtained from the RAP-modulated fluid process, when restricted to the original Lévy process, satisfies the defining Laplace transform identity (ψ(·) - q)^{-1} lacks independent verification for general matrix-exponential jumps. While the reduction to the phase-type case (where it becomes a Markov-modulated Brownian motion) is noted, no separate analytic confirmation or numerical test is provided that the embedding commutes with the resolvent operator outside this subclass. This is central to the extension and requires additional justification.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive assessment of our manuscript and the detailed feedback on the central construction. We address the major comment below.

read point-by-point responses

  1. Referee: The claim that the scale function obtained from the RAP-modulated fluid process, when restricted to the original Lévy process, satisfies the defining Laplace transform identity (ψ(·) - q)^{-1} lacks independent verification for general matrix-exponential jumps. While the reduction to the phase-type case (where it becomes a Markov-modulated Brownian motion) is noted, no separate analytic confirmation or numerical test is provided that the embedding commutes with the resolvent operator outside this subclass. This is central to the extension and requires additional justification.

    Authors: We agree that an explicit independent verification of the Laplace transform identity for general matrix-exponential jumps would strengthen the central claim. In the revised manuscript we will add a dedicated subsection deriving the identity directly from the RAP embedding. The argument proceeds by showing that the resolvent of the fluid process, when projected onto the original Lévy process via the orbit map, coincides with (ψ(s) - q)^{-1}; this uses the fact that the RAP modulation preserves the infinitesimal generator on the relevant subspace and the exit-time distributions for the spectrally negative process. The derivation does not rely on the phase-type reduction. We will also include a numerical check for a concrete matrix-exponential distribution that is not phase-type, comparing the explicit formula against numerical inversion of the Laplace transform. revision: yes

Circularity Check

0 steps flagged

No circularity: derivation proceeds from Laplace-transform definition via independent embedding construction

full rationale

The paper starts from the standard Laplace-transform characterization of the q-scale function as the inverse of (ψ(·) − q) and constructs an explicit matrix representation by embedding the spectrally negative Lévy process into a RAP-modulated fluid process. This embedding is introduced as a new technical device that extends the phase-type case of Ivanovs (2021) without any fitted parameters, self-referential equations, or load-bearing self-citations; the resulting iterative schemes are derived directly from the fluid-process resolvent and reduce to the known phase-type formulas when restricted to that subclass. Because the central claim is obtained by algebraic manipulation of the embedding rather than by re-expressing the target scale function in terms of itself, the derivation chain remains self-contained and externally verifiable against the defining Laplace identity.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central construction rests on the existence of a RAP-modulated fluid embedding that reproduces the resolvent of the original Lévy process. No free parameters are introduced. The background theory of Lévy processes, scale functions, and matrix-exponential distributions is taken from prior literature.

axioms (2)
  • standard math The Laplace exponent ψ of a spectrally negative Lévy process admits the representation that allows the scale function to be recovered from the inverse of (ψ(·)−q).
    Standard definition invoked in the first sentence of the abstract.
  • domain assumption A rational arrival process (RAP) can be used to modulate a stochastic fluid process whose exit probabilities coincide with those of the original Lévy process.
    This is the key modeling step that replaces the phase-type embedding; it is stated as the method that overcomes the limitation of Ivanovs' approach.

pith-pipeline@v0.9.0 · 5533 in / 1419 out tokens · 35646 ms · 2026-05-10T17:53:26.587872+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

44 extracted references · 44 canonical work pages

  1. [1]

    Russian and A merican put options under exponential phase-type L \'e vy models

    S ren Asmussen, Florin Avram, and Martijn R Pistorius. Russian and A merican put options under exponential phase-type L \'e vy models. Stochastic Processes and their Applications , 109(1):79--111, 2004

    work page 2004

  2. [2]

    Renewal theory and queueing algorithms for matrix-exponential distributions

    S ren Asmussen and Mogens Bladt. Renewal theory and queueing algorithms for matrix-exponential distributions. Lecture notes in pure and applied mathematics , pages 313--342, 1996

    work page 1996

  3. [3]

    Point processes with finite-dimensional conditional probabilities

    S ren Asmussen and Mogens Bladt. Point processes with finite-dimensional conditional probabilities. Stochastic Processes and their Applications , 82(1):127--142, 1999

    work page 1999

  4. [4]

    From PH / MAP to ME / RAP

    S ren Asmussen and Mogens Bladt. From PH / MAP to ME / RAP . Queueing Systems , 100(3):173--175, 2022

    work page 2022

  5. [5]

    Dini derivatives and regularity for exchangeable increment processes

    Osvaldo Angtuncio Hern\' a ndez and Ger\' o nimo Uribe Bravo. Dini derivatives and regularity for exchangeable increment processes. Trans. Amer. Math. Soc. Ser. B , 7:24--45, 2020

    work page 2020

  6. [6]

    Avram, A

    F. Avram, A. E. Kyprianou, and M. R. Pistorius. Exit problems for spectrally negative L \'evy processes and applications to ( C anadized) R ussian options. Ann. Appl. Probab. , 14(1):215--238, 2004

    work page 2004

  7. [7]

    Pistorius

    Florin Avram, Zbigniew Palmowski, and Martijn R. Pistorius. On the optimal dividend problem for a spectrally negative L \'evy process. The Annals of Applied Probability , 17:156--180, 2007

    work page 2007

  8. [8]

    L \'e vy processes, phase-type distributions, and martingales

    S ren Asmussen. L \'e vy processes, phase-type distributions, and martingales. Stochastic Models , 30(4):443--468, 2014

    work page 2014

  9. [9]

    An extension of P itman's theorem for spectrally positive L \'evy processes

    Jean Bertoin. An extension of P itman's theorem for spectrally positive L \'evy processes. Ann. Probab. , 20(3):1464--1483, 1992

    work page 1992

  10. [10]

    L \'e vy processes , volume 121 of Cambridge Tracts in Mathematics

    Jean Bertoin. L \'e vy processes , volume 121 of Cambridge Tracts in Mathematics . Cambridge University Press, Cambridge, 1996

    work page 1996

  11. [11]

    N. H. Bingham. Continuous branching processes and spectral positivity. Stochastic Process. Appl. , 4(3):217--242, 1976

    work page 1976

  12. [12]

    Quasi-birth-and-death processes with rational arrival process components

    Nigel G Bean and Bo Friis Nielsen. Quasi-birth-and-death processes with rational arrival process components. Stochastic Models , 26(3):309--334, 2010

    work page 2010

  13. [13]

    Matrix-Exponential Distributions in Applied Probability

    Mogens Bladt and Bo Friis Nielsen. Matrix-Exponential Distributions in Applied Probability . Springer, 2017

    work page 2017

  14. [14]

    Bean, Giang T

    Nigel G. Bean, Giang T. Nguyen, Bo F. Nielsen, and Oscar Peralta. RAP -modulated fluid processes: First passages and the stationary distribution. Stochastic Processes and their Applications , 149:308--340, 2022

    work page 2022

  15. [15]

    Algorithms for return probabilities for stochastic fluid flows

    Nigel G Bean, Ma gorzata M O'Reilly, and Peter G Taylor. Algorithms for return probabilities for stochastic fluid flows. Stochastic Models , 21(1):149--184, 2005

    work page 2005

  16. [16]

    Minimal representations for matrix-exponential distributions

    Peter Buchholz and Mikl \'o s Telek. Minimal representations for matrix-exponential distributions. Queueing Systems , 73:395--427, 2013

    work page 2013

  17. [17]

    Chaumont and R

    L. Chaumont and R. A. Doney. On L \'evy processes conditioned to stay positive. Electron. J. Probab. , 10:no. 28, 948--961, 2005

    work page 2005

  18. [18]

    David R. Cox. A use of complex probabilities in the theory of stochastic processes. Proceedings of the Cambridge Philosophical Society , 51:313--319, 1955

    work page 1955

  19. [19]

    Two-sided reflection of M arkov-modulated B rownian motion

    B D'Auria, J Ivanovs, O Kella, and M Mandjes. Two-sided reflection of M arkov-modulated B rownian motion. Stochastic Models , 28(2):316--332, 2012

    work page 2012

  20. [20]

    Phase-type fitting of scale functions for spectrally negative L \'e vy processes

    Masahiko Egami and Kazutoshi Yamazaki. Phase-type fitting of scale functions for spectrally negative L \'e vy processes. Journal of Computational and Applied Mathematics , 264:1--22, 2014

    work page 2014

  21. [21]

    An introduction to probability theory and its applications

    William Feller. An introduction to probability theory and its applications. V ol. II . Second edition. John Wiley & Sons Inc., New York, 1971

    work page 1971

  22. [22]

    Gerber and Elias S

    Hans U. Gerber and Elias S. W. Shiu. On the time value of ruin. N. Am. Actuar. J. , 2(1):48--78, 1998. With discussion and a reply by the authors

    work page 1998

  23. [23]

    Numerical inverse L aplace transformation using concentrated matrix exponential distributions

    G \'a bor Horv \'a th, Ill \'e s Horv \'a th, Salah Al-Deen Almousa, and Mikl \'o s Telek. Numerical inverse L aplace transformation using concentrated matrix exponential distributions. Performance Evaluation , 137:102067, 2020

    work page 2020

  24. [24]

    High order concentrated matrix-exponential distributions

    G \'a bor Horv \'a th, Ill \'e s Horv \'a th, and Mikl \'o s Telek. High order concentrated matrix-exponential distributions. Stochastic Models , 36(2):176--192, 2020

    work page 2020

  25. [25]

    Concentrated matrix exponential distributions

    Ill \'e s Horv \'a th, Orsolya S \'a f \'a r, Mikl \'o s Telek, and Bence Z \'a mb \'o . Concentrated matrix exponential distributions. In European Workshop on Performance Engineering , pages 18--31. Springer, 2016

    work page 2016

  26. [26]

    Occupation densities in solving exit problems for M arkov additive processes and their reflections

    Jevgenijs Ivanovs and Zbigniew Palmowski. Occupation densities in solving exit problems for M arkov additive processes and their reflections. Stochastic Process. Appl. , 122(9):3342--3360, 2012

    work page 2012

  27. [27]

    Markov-modulated B rownian motion with two reflecting barriers

    Jevgenijs Ivanovs. Markov-modulated B rownian motion with two reflecting barriers. Journal of Applied Probability , 47(4):1034--1047, 2010

    work page 2010

  28. [28]

    On scale functions for L \'e vy processes with negative phase-type jumps

    Jevgenijs Ivanovs. On scale functions for L \'e vy processes with negative phase-type jumps. Queueing Systems , 99:1--28, 2021

    work page 2021

  29. [29]

    Kyprianou, and Victor Rivero

    Alexey Kuznetsov, Andreas E. Kyprianou, and Victor Rivero. The theory of scale functions for spectrally negative L \'evy processes. In L\'evy Matters II , volume 2061 of Lecture Notes in Mathematics . Springer, 2013

    work page 2061

  30. [30]

    Kyprianou

    Andreas E. Kyprianou. Fluctuations of L \'evy Processes with Applications . Springer, 2 edition, 2014

    work page 2014

  31. [31]

    W iener- H opf factorization for L \'e vy processes having positive jumps with rational transforms

    Alan L Lewis and Ernesto Mordecki. W iener- H opf factorization for L \'e vy processes having positive jumps with rational transforms. Journal of Applied Probability , 45(1):118--134, 2008

    work page 2008

  32. [32]

    P. W. Millar. Zero-one laws and the minimum of a M arkov process. Trans. Amer. Math. Soc. , 226:365--391, 1977

    work page 1977

  33. [33]

    Concentrated matrix exponential distributions with real eigenvalues

    Andr \'a s M \'e sz \'a ros and Mikl \'o s Telek. Concentrated matrix exponential distributions with real eigenvalues. Probability in the Engineering and Informational Sciences , 36(4):1171--1187, 2022

    work page 2022

  34. [34]

    A versatile M arkovian point process

    Marcel F Neuts. A versatile M arkovian point process. Journal of Applied Probability , 16(4):764--779, 1979

    work page 1979

  35. [35]

    Marcel F. Neuts. Matrix-Geometric Solutions in Stochastic Models: An Algorithmic Approach . Johns Hopkins University Press, 1981

    work page 1981

  36. [36]

    Nguyen and Oscar Peralta

    Giang T. Nguyen and Oscar Peralta. Rate of strong convergence to M arkov-modulated B rownian motion. Journal of Applied Probability , 59(1):1--16, 2022

    work page 2022

  37. [37]

    Characterization of phase-type distributions

    Colm Art O'Cinneide. Characterization of phase-type distributions. Stochastic Models , 6(1):1--57, 1990

    work page 1990

  38. [38]

    A M arkov jump process associated with the matrix-exponential distribution

    Oscar Peralta. A M arkov jump process associated with the matrix-exponential distribution. Journal of Applied Probability , 60(1):1--13, 2023

    work page 2023

  39. [39]

    Rational arrival processes with strictly positive densities need not be M arkovian

    Oscar Peralta. Rational arrival processes with strictly positive densities need not be M arkovian. arXiv preprint arXiv:2603.28047 , 2026

    work page arXiv 2026

  40. [40]

    Pistorius

    Martijn R. Pistorius. A potential-theoretical review of some exit problems of spectrally negative L \'evy processes. In S\'eminaire de P robabilit\'es XXXVIII , volume 1857 of Lecture Notes in Math. , pages 30--41. Springer, Berlin, 2005

    work page 2005

  41. [41]

    On maxima and ladder processes for a dense class of L \'e vy process

    Martijn Pistorius. On maxima and ladder processes for a dense class of L \'e vy process. Journal of Applied Probability , 43(1):208--220, 2006

    work page 2006

  42. [42]

    B. A. Rogozin. The local behavior of processes with independent increments. Teor. Verojatnost. i Primenen. , 13:507--512, 1968

    work page 1968

  43. [43]

    L. C. G. Rogers. The two-sided exit problem for spectrally positive L \'evy processes. Adv. in Appl. Probab. , 22(2):486--487, 1990

    work page 1990

  44. [44]

    The two-matrix problem

    Mikl \'o s Telek. The two-matrix problem. Queueing Systems , 100(3):265--267, 2022

    work page 2022