pith. sign in

arxiv: 2605.15395 · v1 · pith:PE5RC6NGnew · submitted 2026-05-14 · 🧮 math.PR

Equivalence and Separation for Multivariate Matrix-Exponential and Phase-Type Distribution Classes

Oscar Peralta This is my paper

Pith reviewed 2026-05-19 15:34 UTC · model grok-4.3

classification 🧮 math.PR
keywords multivariate matrix-exponential distributionsmultivariate phase-type distributionsKulkarni representationfactorization conditionWishart trace distributionprojection classesalgebraic classesrational Laplace transforms
0
0 comments X

The pith

The projection-defined multivariate matrix-exponential class equals Kulkarni's algebraic class, while the phase-type inclusion is strict from the trivariate case onward.

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

The paper shows that in the matrix-exponential setting, the class MVME defined via projections of finite-state Markov processes is identical to the algebraic class MME* introduced by Kulkarni. The proof uses a multivariate state-space realization theorem together with elementary state augmentations to reach the Kulkarni form. In the phase-type setting the paper proves that the algebraic subclass MPH* is strictly smaller than the projection-based class MVPH once the dimension reaches three or more. The separation rests on a factorization condition that had not been stated before, together with an explicit counterexample given by the trace of a Wishart matrix, which lies in MVPH but violates the condition. Readers care because these equivalences and separations determine which multivariate lifetime models admit finite-dimensional representations and which density or support features are possible.

Core claim

In the matrix-exponential case, the projection-defined class MVME coincides with Kulkarni's algebraic class MME*. In the phase-type setting, the inclusion of MPH* in MVPH is strict from the trivariate case onward, shown by a factorization condition and a Wishart trace distribution that belongs to MVPH but fails the condition.

What carries the argument

Multivariate state-space realization theorem combined with elementary augmentations to reach Kulkarni form, plus a new factorization condition that characterizes the algebraic phase-type subclass MPH*.

If this is right

  • Every proper rational multivariate Laplace transform admits a finite-dimensional Kulkarni-type representation once Markovian sign constraints are removed.
  • Projection-based multivariate phase-type distributions can possess density shapes and support geometries that do not appear in the classical univariate theory.
  • The strict separation between the algebraic and projection classes begins precisely at the trivariate level.

Where Pith is reading between the lines

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

  • The equivalence result may allow existing univariate algorithms for matrix-exponential fitting to be lifted directly to the multivariate setting.
  • The factorization condition supplies a practical test that could be used to decide whether a given multivariate phase-type law admits an algebraic representation.
  • Models in reliability or risk analysis that rely on multivariate phase-type margins may now be checked for membership in the smaller algebraic class before simulation.

Load-bearing premise

The multivariate state-space realization theorem can be combined with elementary augmentations to obtain a Kulkarni-type representation, and the factorization condition correctly separates MPH* from the larger MVPH class.

What would settle it

A direct verification showing whether the trace of a Wishart matrix satisfies the stated factorization condition while still admitting a projection representation in MVPH.

read the original abstract

We resolve two questions left open by Bladt and Nielsen (2010) concerning multivariate families of matrix-exponential and phase-type distributions. First, in the matrix-exponential case, the projection-defined class MVME coincides with Kulkarni's algebraic class MME*. Our proof combines a multivariate state-space realization theorem with elementary augmentations that put the realization into Kulkarni's form. Thus every proper rational multivariate Laplace transform has a finite-dimensional Kulkarni-type representation once Markovian sign constraints are removed. Second, in the phase-type setting, the inclusion of MPH* in MVPH is strict from the trivariate case onward. The separation is obtained through a factorization condition for MPH* that appears not to have been previously identified in the PH literature. A Wishart trace distribution belongs to MVPH but fails this condition, hence providing the required example outside MPH*. The example also shows that projection-based multivariate phase-type laws may have density and support geometry that are absent from the usual univariate theory.

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 paper resolves two open questions from Bladt and Nielsen (2010) on multivariate matrix-exponential and phase-type distributions. It proves that the projection-defined MVME class coincides with Kulkarni's algebraic MME* class by combining a multivariate state-space realization theorem with elementary augmentations to produce a Kulkarni-type representation. For the phase-type setting, it establishes that the inclusion MPH* ⊂ MVPH is strict from the trivariate case onward via a newly identified factorization condition that characterizes MPH*, together with an explicit Wishart trace distribution that lies in MVPH but violates the condition.

Significance. If the constructions and counterexample hold, the results close important gaps in the classification of multivariate ME and PH families, clarifying the effect of removing Markovian sign constraints and identifying a concrete separation via the factorization condition. The explicit state-space augmentations and the Wishart-trace example (which also illustrates non-standard density and support geometry) provide verifiable, constructive evidence rather than abstract existence arguments, strengthening the foundation for applications in multivariate stochastic modeling.

major comments (2)
  1. The equivalence proof relies on combining the multivariate state-space realization theorem with elementary augmentations to reach Kulkarni form; the manuscript should explicitly verify in the relevant theorem or proposition that these augmentations preserve the rational Laplace transform without introducing extraneous parameters or sign constraints.
  2. The separation result hinges on the Wishart trace distribution belonging to MVPH while failing the factorization condition for MPH*; the paper must supply the explicit computation (Laplace transform or moment-generating function) confirming violation of the condition, as this is load-bearing for the strict-inclusion claim from the trivariate case onward.
minor comments (2)
  1. The introduction should include the full bibliographic details for the cited Bladt and Nielsen (2010) reference.
  2. Notation for the classes (MVME, MME*, MVPH, MPH*) is introduced clearly but would benefit from a consolidated table or diagram summarizing the inclusions and equivalences.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading, the positive assessment of the results, and the recommendation for minor revision. The two major comments identify places where additional explicit verification will strengthen the presentation. We respond to each comment below and will incorporate the suggested clarifications in the revised manuscript.

read point-by-point responses

  1. Referee: The equivalence proof relies on combining the multivariate state-space realization theorem with elementary augmentations to reach Kulkarni form; the manuscript should explicitly verify in the relevant theorem or proposition that these augmentations preserve the rational Laplace transform without introducing extraneous parameters or sign constraints.

    Authors: We agree that an explicit verification step will improve clarity. In the revised manuscript we will add a short auxiliary proposition immediately after the statement of the main equivalence result. The proposition will compute the Laplace transform of the augmented representation explicitly, confirming that it coincides with the original rational transform. The augmentations are realized by block-matrix embeddings and coordinate permutations; these operations are purely algebraic, introduce no new scalar parameters, and impose no additional sign constraints beyond those already present in the underlying state-space realization. revision: yes

  2. Referee: The separation result hinges on the Wishart trace distribution belonging to MVPH while failing the factorization condition for MPH*; the paper must supply the explicit computation (Laplace transform or moment-generating function) confirming violation of the condition, as this is load-bearing for the strict-inclusion claim from the trivariate case onward.

    Authors: We accept that the explicit verification is necessary to make the separation fully self-contained. Although the manuscript already identifies the factorization condition and asserts that the Wishart-trace law violates it, the revised version will include the complete derivation of the multivariate Laplace transform of the trace distribution together with the direct algebraic check that the resulting expression fails the factorization condition. This computation will be placed in the section containing the counterexample and will remain restricted to the trivariate case, thereby supporting the strict-inclusion statement without altering any other claims. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The central claims are established through an external multivariate state-space realization theorem combined with explicit elementary augmentations to reach Kulkarni's algebraic form for the MVME=MME* equivalence, plus a newly derived factorization condition for MPH* together with a concrete Wishart trace distribution counterexample that lies in MVPH but violates the condition. These steps rely on direct constructions, verifiable distributions, and independent prior results rather than any self-definitional reduction, fitted input renamed as prediction, or load-bearing self-citation chain. The derivation remains self-contained against external benchmarks with independent mathematical content.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claims rest on the multivariate state-space realization theorem and a newly stated factorization condition for MPH*, with no free parameters or invented entities introduced.

axioms (2)
  • domain assumption Multivariate state-space realization theorem holds and permits elementary augmentations to Kulkarni form
    Invoked directly in the proof that MVME coincides with MME*.
  • ad hoc to paper Factorization condition characterizes membership in MPH*
    Newly identified condition used to separate MPH* from MVPH.

pith-pipeline@v0.9.0 · 5694 in / 1394 out tokens · 64023 ms · 2026-05-19T15:34:02.193898+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [1]

    Daniel Alpay and C. Dubi. A realization theorem for rational functions of several complex variables.Systems & Control Letters, 49(3):225–229, 2003

    work page 2003

  2. [2]

    Springer, New York, 2 edition, 2003

    Søren Asmussen.Applied Probability and Queues, volume 51 ofStochastic Modelling and Applied Probability. Springer, New York, 2 edition, 2003. 18

    work page 2003

  3. [3]

    Renewal theory and queueing algorithms for matrix- exponential distributions

    Søren Asmussen and Mogens Bladt. Renewal theory and queueing algorithms for matrix- exponential distributions. In S. R. Chakravarthy and A. S. Alfa, editors,Matrix-Analytic Methods in Stochastic Models, volume 183 ofLecture Notes in Pure and Applied Mathematics, pages 313–341. Marcel Dekker, New York, 1997

    work page 1997

  4. [4]

    Langberg, Thomas H

    David Assaf, Naftali A. Langberg, Thomas H. Savits, and Moshe Shaked. Multivariate phase- type distributions.Operations Research, 32(3):688–702, 1984

    work page 1984

  5. [5]

    Rap-modulated fluid processes: First passages and the stationary distribution.Stochastic Processes and their Ap- plications, 149:308–340, 2022

    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 Ap- plications, 149:308–340, 2022

    work page 2022

  6. [6]

    Plemmons.Nonnegative Matrices in the Mathematical Sci- ences

    Abraham Berman and Robert J. Plemmons.Nonnegative Matrices in the Mathematical Sci- ences. SIAM, Philadelphia, 1994

    work page 1994

  7. [7]

    A tractable class of multivariate phase-type distributions for loss modeling

    Martin Bladt. A tractable class of multivariate phase-type distributions for loss modeling. North American Actuarial Journal, 27(4):710–730, 2023

    work page 2023

  8. [8]

    Assessing continuous common-shock risk through matrix distributions.Scandinavian Actuarial Journal, 2026

    Martin Bladt, Oscar Peralta, and Jorge Yslas. Assessing continuous common-shock risk through matrix distributions.Scandinavian Actuarial Journal, 2026. Advance online pub- lication

    work page 2026

  9. [9]

    Mogens Bladt and Marcel F. Neuts. Matrix-exponential distributions: Calculus and interpre- tations via flows.Stochastic Models, 19(1):113–124, 2003

    work page 2003

  10. [10]

    Multivariate matrix-exponential distributions.Stochastic Models, 26(1):1–26, 2010

    Mogens Bladt and Bo Friis Nielsen. Multivariate matrix-exponential distributions.Stochastic Models, 26(1):1–26, 2010

    work page 2010

  11. [11]

    Springer, New York, 2017

    Mogens Bladt and Bo Friis Nielsen.Matrix-Exponential Distributions in Applied Probability, volume 81 ofProbability Theory and Stochastic Modelling. Springer, New York, 2017

    work page 2017

  12. [12]

    Parisian types of ruin probabilities for a class of dependent risk-reserve processes.Scandinavian Actuarial Journal, 2019(1):32–61, 2019

    Mogens Bladt, Bo Friis Nielsen, and Oscar Peralta. Parisian types of ruin probabilities for a class of dependent risk-reserve processes.Scandinavian Actuarial Journal, 2019(1):32–61, 2019

    work page 2019

  13. [13]

    Realization of multidimensional systems in fornasini-marchesini state-space model.Multidimensional Systems and Signal Pro- cessing, 22(4):319–333, 2011

    Hua Cheng, Tatsuya Saito, Shin-ya Matsushita, and Li Xu. Realization of multidimensional systems in fornasini-marchesini state-space model.Multidimensional Systems and Signal Pro- cessing, 22(4):319–333, 2011

    work page 2011

  14. [14]

    Eric C. K. Cheung, Oscar Peralta, and Jae-Kyung Woo. Multivariate matrix-exponential affine mixtures and their applications in risk theory.Insurance: Mathematics and Economics, 106:364–389, 2022

    work page 2022

  15. [15]

    Horn, and Vladimir V

    Vyacheslav Futorny, Roger A. Horn, and Vladimir V. Sergeichuk. Classification of squared nor- mal operators on unitary and euclidean spaces.Journal of Mathematical Sciences, 155(6):950– 955, 2008

    work page 2008

  16. [16]

    Gupta and Daya K

    Arjun K. Gupta and Daya K. Nagar.Matrix Variate Distributions. Chapman & Hall/CRC, Boca Raton, 2000. 19

    work page 2000

  17. [17]

    Horn and Charles R

    Roger A. Horn and Charles R. Johnson.Matrix Analysis. Cambridge University Press, Cam- bridge, 2 edition, 2013

    work page 2013

  18. [18]

    Kulkarni

    Vidyadhar G. Kulkarni. A new class of multivariate phase type distributions.Operations Research, 37(1):151–158, 1989

    work page 1989

  19. [19]

    Neuts.Matrix-Geometric Solutions in Stochastic Models: An Algorithmic Approach

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

    work page 1981

  20. [20]

    A Markov jump process associated with the matrix-exponential distribution

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

    work page 2023

  21. [21]

    Ruin problems for risk processes with dependent phase- type claims.Methodology and Computing in Applied Probability, 25(4):86, 2023

    Oscar Peralta and Matthieu Simon. Ruin problems for risk processes with dependent phase- type claims.Methodology and Computing in Applied Probability, 25(4):86, 2023

    work page 2023

  22. [22]

    The generalised product moment distribution in samples from a normal multi- variate population.Biometrika, 20A(1–2):32–52, 1928

    John Wishart. The generalised product moment distribution in samples from a normal multi- variate population.Biometrika, 20A(1–2):32–52, 1928. 20

    work page 1928