Pith. sign in

REVIEW 4 major objections 4 minor 39 references

Positive Bidiagonal Factorizations for Banded Markov Processes

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read An ordered positive bidiagonal factorization is a full spectral and probabilistic description of a banded Markov chain, not just a certificate of total nonnegativity.

desk verdict A substantial, mostly self-contained extension of Mañas's PBF spectral program, with the central Karlin–McGregor representation inherited from two prior papers whose hypotheses are never stated. read the letter →

arxiv 2608.00788 v1 pith:S4JL7GGU submitted 2026-08-01 math.CA math-phmath.MPmath.PRmath.SP

Manuel Mañas This is my paper · ORCID
classification math.CAmath-phmath.MPmath.PRmath.SP MSC 15B4842C0560J1060J2747B36
keywords positivebidiagonalfactorizationtotalpositivitybandedMarkovprocessesmixed-typemultipleorthogonalityKarlin–McGregorrepresentationmatrixcontinuedfractionsPiñeiropolynomialsJacobi-likeweights
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 tries to establish that an ordered positive bidiagonal factorization (PBF) of a banded Markov transition matrix carries the entire spectral and probabilistic structure of the chain. The same factors that certify total nonnegativity are shown to determine two families of mixed-type multiple orthogonal polynomials, an entrywise positive matrix of measures, and an ordered sequence of elementary death-or-stay and birth-or-stay Markov transitions. From these, the paper derives Karlin–McGregor formulas for transition probabilities, Green kernels, resolvents, potentials, and first-passage transforms without requiring reversibility or block symmetrization. It also characterizes rational stochastic PBFs by ordered finite-urn experiments, links cyclic reorderings to Darboux transformations, and obtains exact positivity classifications for the mixed Piñeiro and Jacobi-like systems. A sympathetic reader would care because this extends the classical birth–death spectral theory to arbitrary bandwidth under one explicit structural hypothesis.

What carries the argument

The central object is the ordered positive stochastic bidiagonal factorization T=M1⋯Md, where every Mi is bidiagonal and positive: lower factors allow only n→n−1 or stay, upper factors only n→n+1 or stay. This one structure simultaneously encodes the elementary Markov transitions, the mixed-type multiple orthogonal polynomials and positive matrix of measures through the Favard theorem, the finite-urn realization through reduced rational probabilities, and the matrix continued fraction for return laws. A Doob-type normalization by the positive harmonic vector h_N=B_N(1) turns the recurrence matrix into the row-stochastic transition kernel without changing the spectral measure.

What would settle it

Take the rational (3,2) Piñeiro chain with α=(−1/3,0,1/3), β=(0,1/3), enumerate the return probabilities to state 0 from the explicit transition matrix (111), and compare the Taylor coefficients of f0(z) with f0(z)=1−1/2F1(1,2/3;5/3;z); a single mismatch at any order would refute the continued-fraction/renewal identity.

Watch

Extended reading notes

Core claim

The central claim is that a banded transition matrix factored as T=L1⋯LpUq⋯U1, with positive lower and upper bidiagonal factors, is more than a totally nonnegative matrix: the ordered factors themselves are the spectral and probabilistic data. Each lower factor is a death-or-stay binary experiment and each upper factor is a birth-or-stay experiment; Theorem 3.1 proves that for rational entries this is exactly an ordered sequence of finite two-color urn experiments. Through the mixed-type Favard theory, the same factors generate two left–right polynomial families and a positive q×p matrix of measures on [0,1], giving explicit Karlin–McGregor formulas for all powers of the kernel, the Green ke

Load-bearing premise

The load-bearing premise is that the earlier Favard theorems ([10] and [11, Theorem 2.19]) and the companion paper's boundary evaluation of the factor-resolved continued fraction [33] apply exactly as stated to every positive stochastic bidiagonal factorization; the paper explicitly says the continued-fraction construction is not reproved here, so if those imported results fail or need narrower hypotheses, the spectral formulas and the first-return law lose support.

Editorial extensions

If this is right

  • For any banded stochastic matrix with a positive stochastic bidiagonal factorization, every transition probability admits the explicit mixed Karlin–McGregor form (9); in bounded continuous time, uniformization merely replaces x^k by e^{-νt(1−x)} and leaves the polynomial families and measure matrix unchanged.
  • Rational stochastic PBFs are exactly ordered finite-urn experiments: each factor is realized by the smallest finite two-color urn determined by its reduced probabilities, and one full cycle of draws realizes one transition of the chain.
  • The complete first-return law from state 0 is determined by the boundary matrix continued fraction of the individual factors via the renewal identity f0(z)=1−1/g0(z); for the rational (3,2) Piñeiro chain this becomes the explicit formula f0(z)=1−1/2F1(1,2/3;5/3;z).
  • Bounded generators built by Poissonization inherit the discrete spectral data: recurrence is decided by divergence of ∫ dψ_{1,1}(x)/(1−x), ergodicity by a mass point at x=1, and endpoint exponents give power-law decay in k or t.
  • For unbounded exit rates, a conservative generator whose shifted leading truncations all admit scalar PBFs must be tridiagonal, but the construction Q=V(T−I) preserves arbitrary fixed bandwidth, is non-explosive when holding rates are linear, and its recurrence and hitting probabilities are those of the embedded chain T.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If Theorem 3.1 is read as a simulation recipe, any rational banded chain with a PBF can be sampled exactly by fixed-order draws from finite urns, which suggests a practical exact Monte Carlo scheme even for large bandwidths and non-reversible wide jumps.
  • The Piñeiro classification that strict cyclic ordering is the only open PBF chamber for q∈{2,3,4} invites the conjecture that cyclic order may be the generic open chamber for every q, with lower-dimensional cancellation strata described by combinatorial conditions; the paper does not assert this for all q.
  • Corollary 6.4 implies that when p≠q one should not search for a block-diagonal symmetrizer to apply classical matrix-valued orthogonal polynomial methods; the two-sided mixed spectral representation is the natural object, potentially guiding numerical algorithms for phase-resolved return probabilities.
  • The state-dependent construction Q=V(T−I) suggests that the embedded chain, not the holding-time clock, controls qualitative long-time behavior; one testable extension is whether recurrence and eventual hitting probabilities remain invariant under arbitrary positive state-dependent time changes of the displayed type.
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.

Referee Report

4 major / 4 minor

Summary. The paper develops a spectral and probabilistic theory for banded Markov transition matrices that admit an ordered positive bidiagonal factorization (PBF). It claims that the individual bidiagonal factors determine two families of mixed-type multiple orthogonal polynomials, an entrywise positive q×p matrix of measures, and a sequence of elementary death-or-stay and birth-or-stay transitions, yielding Karlin–McGregor formulas for transition probabilities, Green kernels, resolvents, potentials, and first-passage transforms without reversibility or block symmetrizability. The paper also gives finite-urn equivalences, Darboux intertwinings by cyclic reordering, a factor-resolved continued fraction for first-return laws, QBD reformulations, bounded and unbounded continuous-time generators, and explicit classifications for the mixed Piñeiro and Jacobi-like systems, including a rational (3,2) example.

Significance. If the main claims hold, the paper gives a genuinely unified framework: the same bidiagonal factorization simultaneously certifies total nonnegativity, supplies spectral measures and orthogonal polynomial families, and provides elementary Markov transitions and continued-fraction return laws. The explicit Piñeiro treatment is a substantial concrete contribution: the Cauchy construction in §7, the exact PBF region in Theorem 7.8, the structural-band positivity regions in Theorem 7.12, and the worked (3,2) example in §8 are all given by finite, explicit formulas and sign counts. The block-symmetrizability obstruction in Corollary 6.4 is a useful and clearly stated result. The paper is weaker where it depends on external results: the key spectral representation and the continued-fraction boundary evaluation are imported from the author's earlier work and are not reproved or even stated with their hypotheses.

major comments (4)
  1. [§3.3, Corollary 3.2 and Eq. (9)] The central Karlin–McGregor representation is proved by the sentence: 'The bounded spectral Favard theorem [10] and its Markov specialization [11, Theorem 2.19] give the two mixed-type polynomial families, the entrywise positive matrix of measures, and formula (9).' The hypotheses of those theorems are not stated, and the paper does not verify that an arbitrary infinite ordered product of positive stochastic bidiagonal factors satisfies them. Since Eqs. (16), (17), (27), (29), and the endpoint criteria of Theorems 4.7–4.9 all inherit this representation, this is a load-bearing gap. The authors should either state the precise hypotheses and prove that every ordered positive stochastic PBF product satisfies them, or give a self-contained proof of (9).
  2. [§3.2 and §6, Eqs. (7)–(8), (39)–(40)] The first-return law (8) and the QBD return matrices (39)–(40) depend on the boundary evaluation of the matrix continued fraction from the companion paper [33]. The paper explicitly says this construction is 'not reproved here' and is 'combined with the renewal identity.' This is acceptable only if the companion result is stated with enough precision to allow independent verification. As written, the reader cannot check the central probabilistic application from the present manuscript. Please state the exact theorem imported from [33], including its hypotheses and convergence conditions, or provide the proof.
  3. [§7.4, Eqs. (97)–(99)] The stochastic normalization factor by factor uses [11, Proposition 2.3] to convert the positive PBF of T^P into a positive stochastic PBF of P^P. The conversion requires the intermediate vectors h^[r] to be strictly positive. In the ordered cyclic PBF region this follows from positivity of the bidiagonal factors, and the harmonic vector is explicit via (63). However, the paper does not state the domain of validity of [11, Proposition 2.3] for infinite matrices. Since the same proposition is used in Theorem 3.1 for the rational urn equivalence, the infinite-matrix hypothesis should be made explicit.
  4. [§9, Theorem 9.4] The proof of the Jacobi-like PBF region invokes Lemma 9.2 and a Gasca–Peña criterion for strict total positivity, and it introduces a 'large-minor obstruction' in Proposition 9.5 whose proof is only sketched ('the term omitting the largest s labels is the unique dominant term'). This asymptotic dominance argument is load-bearing for the claimed maximality in Proposition 9.5 and Corollaries 9.6–9.8. Please provide the full error estimates or a complete proof of the dominance claim.
minor comments (4)
  1. [Definition 5.3] The phrase 'withinfn vn >0' is a typo; it should read 'with inf_n v_n >0'.
  2. [§2.1.2, Definition 2.2] The notation L^{(1)} ... L^{(p)} with upper indices is clear, but in §3.1 and later the same factors are written M_1,...,M_d with a type word ε. A short table linking the two notations would improve readability.
  3. [§8.3, Eq. (117)] In Proposition 8.5, the claim that (P^P)^k_{0,0} = ϑ/(k+ϑ) is a striking closed form. It would help to note explicitly that this uses the normalization choices A_0^{(1)} = ϑ and B_0^{(1)} = 1 under the Piñeiro spectral measure, so the reader does not confuse this with the generic normalization in Corollary 3.2.
  4. [§5.3, Eq. (27)] The Abel limit lim_{r↑1} is introduced but the spectral measure ψ^{(λ)} is not defined in the text. Please define it precisely, including the normalization of A^{(a,λ)} and B^{(b,λ)}.

Circularity Check

2 steps flagged · score 4.0 of 10

Central KM representation and continued-fraction boundary evaluation are imported from same-author prior work; no definitional circularity, but self-citation is load-bearing.

  1. self citation load bearing [Corollary 3.2, Section 3.3, formula (9)]
    "The bounded spectral Favard theorem [10] and its Markov specialization [11, Theorem 2.19] give the two mixed-type polynomial families, the entrywise positive matrix of measures, and formula (9)."

    The paper's central Karlin–McGregor representation (9) for an ordered positive stochastic bidiagonal factorization is not proved here. The proof reduces to a citation of the author's earlier Favard theorem and its Markov specialization; the introduction already attributes 'a discrete Karlin–McGregor formula for banded Markov kernels with a PBF' to [11]. Thus Corollary 3.2 is, at the proof level, a restatement of the imported theorem rather than a derivation from the urn construction. The hypotheses of the cited theorem are not stated or checked, and the general spectral claim inherits its entire support from that same-author citation.

  2. self citation load bearing [Section 3.2, equations (7)–(8); Section 6, equations (39)–(40)]
    "The continued-fraction construction of [33] is not reproved here: its boundary evaluation is combined with the renewal identity to obtain the complete scalar first-return law (8)."

    The first-return law (8) is obtained by taking the boundary evaluation g_0(τ^M)=e_1^T Φ_0^T(τ)e_1 from the companion paper [33] as an input and then applying the renewal identity. The same imported boundary evaluation underlies the QBD return matrices (39)–(40). The paper explicitly declines to prove that evaluation, so the general claim that the factor-resolved continued fraction gives the first-return law reduces to an unproved identity in a same-author companion paper. The later Piñeiro check (124)–(125) uses the same imported identity to identify the continued fraction with an explicitly computed hypergeometric g_0, so it does not independently establish the general boundary evaluation.

full rationale

Most of the concrete content is self-contained algebra: the Piñeiro factor entries are derived in Appendix A; Theorem 7.8 proves the exact PBF region from the explicit formulas; Theorem 7.12 and Proposition 7.17 use direct determinant/sign arguments; Proposition 5.1, Corollary 6.4, Proposition 8.5, and Proposition 8.13 are proved internally. I find no fitted-input-called-prediction and no definitional equivalence in which an output is identical to an input by construction: no parameter is fitted to a target quantity and renamed a prediction, and the Piñeiro spectral measure is the defining measure of the biorthogonal system, consistently used as the kernel in the explicitly evaluated integrals. The general spectral representation (9) and the continued-fraction boundary evaluation (7) are, however, load-bearing imports from the author's own prior work ([10], [11, Theorem 2.19], [33]), with the latter explicitly 'not reproved here.' Some Jacobi-like normalizer formulas are likewise attributed to [9]. These are self-citations rather than demonstrated circular reductions, but they are not machine-checked or independently verified inside the paper, and they support the central general claims. The paper's independent explicit-model contributions keep the circularity score at 4 rather than higher.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities and fits no parameters to data; the parameters alpha, beta, a, b are model inputs. The main external assumptions are the spectral Favard theorems from the author's prior work and the continued fraction boundary evaluation from the companion paper [33].

assumptions (5)
  • domain assumption A matrix with a positive bidiagonal factorization can be converted, preserving positivity, into a positive stochastic bidiagonal factorization (from [11, Proposition 2.3]).
    Used in Theorem 3.1 to move from rational PBF to urn probabilities; not reproved in this paper.
  • domain assumption The bounded spectral Favard theorem applies to banded PBF matrices and yields two families of mixed-type multiple orthogonal polynomials and an entrywise positive matrix of measures ([10] and [11, Theorem 2.19]).
    Invoked in Corollary 3.2 and throughout Sections 4 and 6 to obtain Karlin-McGregor formulas.
  • domain assumption The boundary evaluation of the matrix continued fraction in [33] is correct and applies to the bidiagonal factors considered here.
    Used in equation (7) and Equations (39)-(40) for first-return and QBD return matrices; the paper explicitly says the construction is not reproved.
  • standard math Cauchy determinant identity and total positivity minor criteria (Gasca-Peña) are valid as stated.
    Used in the Piñeiro Cauchy moment matrix formulas and in the Jacobi-like strict total positivity arguments.
  • standard math The Lyapunov criterion for non-explosion of continuous-time Markov chains is valid.
    Used in Proposition 5.5 and in the local-clock constructions for unbounded generators.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Positive Bidiagonal Factorizations for Banded Markov Processes." pith.science (2026). https://pith.science/paper/S4JL7GGU

@misc{pith2026260800788,
  author       = {Pith},
  title        = {Pith review of: Positive Bidiagonal Factorizations for Banded Markov Processes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/S4JL7GGU}},
  note         = {Machine review of arXiv:2608.00788}
}
abstract

An ordered positive bidiagonal factorization (PBF) is used to develop a spectral and probabilistic theory for Markov transition matrices of arbitrary finite bandwidth. The factors determine two families of mixed-type multiple orthogonal polynomials, an entrywise positive $q\times p$ matrix of measures, and a sequence of elementary death-or-stay and birth-or-stay transitions. This yields Karlin-McGregor formulas for transition probabilities, Green kernels, resolvents, potentials, and first-passage transforms without reversibility or block symmetrizability. Rational stochastic PBFs are characterized by ordered finite-urn experiments. Cyclic reorderings give Darboux intertwinings, while a factor-resolved continued fraction gives the first-return law. Grouping states produces a finite-phase quasi-birth-and-death process and a matrix continued fraction for return time and phase. For bounded continuous-time generators, uniformization preserves the spectral data. For unbounded rates, a conservative generator whose shifted leading truncations all admit scalar PBFs must be tridiagonal; nevertheless, $Q=V(T-I)$ preserves arbitrary fixed bandwidth and separates the embedded chain from the holding rates. The theory is explicit for mixed Pi\~neiro and Jacobi-like systems. For Pi\~neiro, the exact PBF region and larger structural-band positivity regions are obtained. Jacobi-like beta-convolution weights admit Gamma-factor cancellations, complete cancellation recovering Pi\~neiro. For $q\in\{2,3,4\}$, the strict ordering conditions give the only open PBF region, with further lower-dimensional cancellation strata. Rational $(3,2)$ examples provide all factors and the resulting Markov models.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 31 canonical work pages

  1. [10]

    AmílcarBranquinho,AnaFoulquié-Moreno,andManuelMañas.Spectralthe- ory for bounded banded matrices with positive bidiagonal factorization and mixed multiple orthogonal polynomials.Advances in Mathematics434(2023) 109313.doi:10.1016/j.aim.2023.109313

    work page arXiv 2023

  2. [33]

    Matrix continued fractions from bidiagonal factorizations

    Manuel Mañas. “Matrix continued fractions from bidiagonal factorizations”. Manuscript in preparation. 2026

    work page 2026

  3. [1]

    Carlos Álvarez-Fernández, Ulises Fidalgo, and Manuel Mañas. Multiple or- thogonal polynomials of mixed type: Gauss–Borel factorization and the multi- component 2D Toda hierarchy.Advances in Mathematics227(2011) 1451– 1525.doi:10.1016/j.aim.2011.03.008

    work page doi:10.1016/j.aim.2011.03.008 2011

  4. [2]

    Disquisitiones analyticae de novo problemate coniecturali

    Daniel Bernoulli. Disquisitiones analyticae de novo problemate coniecturali. Novi Commentarii Academiae Scientiarum Imperialis Petropolitanae14 (1769) 3–25

    work page

  5. [3]

    Amílcar Branquinho, Juan E. F. Díaz, Ana Foulquié-Moreno, and Manuel Mañas. Uniform multiple orthogonal polynomials and Markov chains.Numer- ical Algorithms102(2026) 1743–1772.doi:10.1007/s11075-025-02195-6

    work page doi:10.1007/s11075-025-02195-6 2026

  6. [4]

    Amílcar Branquinho, Juan E. F. Díaz, Ana Foulquié-Moreno, Manuel Mañas, and Carlos Álvarez-Fernández. Jacobi-Piñeiro random walks.Revista de la Real Academia de Ciencias Exactas, Físicas y Naturales. Serie A. Matemáti- cas118(2024) 15.doi:10.1007/s13398-023-01510-x. 94 REFERENCES

    work page doi:10.1007/s13398-023-01510-x 2024

  7. [5]

    Piñeiro mixed orthogonal polynomials and bidiagonal factorization

    Amílcar Branquinho, Juan E. F. Díaz, Ana Foulquié-Moreno, Manuel Mañas, and Thomas Wolfs. “Piñeiro mixed orthogonal polynomials and bidiagonal factorization”. Manuscript in preparation. 2026

    work page 2026

  8. [6]

    Banded totally positive matrices and normality for mixed multiple orthogonal polynomials

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas.Banded to- tally positive matrices and normality for mixed multiple orthogonal polynomi- als. arXiv:2404.13965 [math.CA]. 2024

    work page Pith review arXiv 2024

Show all 39 references
  1. [7]

    Bidiagonal factorization of banded recursion matrices for mixed-type multiple orthogonal polynomials.Linear Algebra and its Applications(2026)

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas. Bidiagonal factorization of banded recursion matrices for mixed-type multiple orthogonal polynomials.Linear Algebra and its Applications(2026). To appear. arXiv: 2603.21345 [math.CA]

    2026

  2. [8]

    Bidiagonal factorization of tetradiagonal matrices and Darboux transformations.Anal- ysis and Mathematical Physics13(2023) 42.doi:10.1007/s13324- 023- 00801-1

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas. Bidiagonal factorization of tetradiagonal matrices and Darboux transformations.Anal- ysis and Mathematical Physics13(2023) 42.doi:10.1007/s13324- 023- 00801-1

    2023 doi

  3. [9]

    Jacobi-like and Laguerre-like mixed-type multiple orthogonal polynomials

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas. “Jacobi-like and Laguerre-like mixed-type multiple orthogonal polynomials”. Manuscript in preparation. 2026

    2026

  4. [11]

    arXiv:2601.10890 [math.PR]

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas.Spectral the- ory for Markov chains with transition matrix admitting a stochastic bidiagonal factorization. arXiv:2601.10890 [math.PR]. 2026

    2026

  5. [12]

    arXiv:2601.12453 [math.CA]

    Amílcar Branquinho, Ana Foulquié-Moreno, and Manuel Mañas.Unbounded banded matrices, shifted positive bidiagonal factorizations, and mixed-type multiple orthogonality. arXiv:2601.12453 [math.CA]. 2026

    2026

  6. [13]

    Pierre Brémaud.Markov Chains. 2nd ed. Vol. 31. Texts in Applied Mathe- matics. Springer, 2020

    2020

  7. [14]

    Studden, and Marcin J

    Holger Dette, Bettina Reuther, William J. Studden, and Marcin J. Zygmunt. Matrixmeasuresandrandomwalkswithablocktridiagonaltransitionmatrix. SIAM Journal on Matrix Analysis and Applications29(2006) 117–142.doi: 10.1137/050638230

    2006 doi

  8. [15]

    Doob.Classical Potential Theory and Its Probabilistic Counter- part

    Joseph L. Doob.Classical Potential Theory and Its Probabilistic Counter- part. Vol. 262. Grundlehren der Mathematischen Wissenschaften. New York: Springer-Verlag, 1984.doi:10.1007/978-1-4612-5208-5

    1984 doi

  9. [16]

    Joseph L. Doob. Conditional Brownian motion and the boundary limits of harmonic functions.Bulletin de la Société Mathématique de France85(1957) 431–458.doi:10.24033/bsmf.1494

    1957 doi

  10. [17]

    Über zwei bekannte Einwände gegen das Boltzmannsche H-Theorem.Physikalische Zeitschrift8 (1907) 311–314

    Paul Ehrenfest and Tatiana Ehrenfest-Afanassjewa. Über zwei bekannte Einwände gegen das Boltzmannsche H-Theorem.Physikalische Zeitschrift8 (1907) 311–314

    1907

  11. [18]

    Fallat and Charles R

    Shaun M. Fallat and Charles R. Johnson.Totally Nonnegative Matrices. Princeton: Princeton University Press, 2011.doi:10.1515/9781400839018

    2011 doi

  12. [19]

    William Feller.An Introduction to Probability Theory and Its Applications, Vol. 2. 2nd ed. New York: John Wiley & Sons, 1971. REFERENCES 95

    1971

  13. [20]

    de la Iglesia

    Lidia Fernández and Manuel D. de la Iglesia. Quasi-birth-and-death processes and multivariate orthogonal polynomials.Journal of Mathematical Analysis and Applications499(2021) 125029.doi:10.1016/j.jmaa.2021.125029

    2021

  14. [21]

    Gantmacher and Mark G

    Felix P. Gantmacher and Mark G. Krein.Oscillation Matrices and Ker- nels and Small Vibrations of Mechanical Systems. Revised. Providence: AMS Chelsea Publishing, 2002.doi:10.1090/chel/345

    2002 doi

  15. [22]

    Total positivity, QR factorization, and Neville elimination.SIAM Journal on Matrix Analysis and Applications 14(1993) 1132–1140.doi:10.1137/0614077

    Mariano Gasca and Juan Manuel Peña. Total positivity, QR factorization, and Neville elimination.SIAM Journal on Matrix Analysis and Applications 14(1993) 1132–1140.doi:10.1137/0614077

    1993 doi

  16. [23]

    Alberto Grünbaum and Manuel D

    F. Alberto Grünbaum and Manuel D. de la Iglesia. An urn model for the Jacobi-Piñeiro polynomials.Proc. Amer. Math. Soc.150(2022) 3613–3625. doi:10.1090/proc/15910

    2022 doi

  17. [24]

    Alberto Grünbaum and Manuel D

    F. Alberto Grünbaum and Manuel D. de la Iglesia. Matrix valued orthog- onal polynomials arising from group representation theory and a family of quasi-birth-and-death processes.SIAM Journal on Matrix Analysis and Ap- plications30(2008) 741–761.doi:10.1137/070697604

    2008 doi

  18. [25]

    Alberto Grünbaum and Manuel D

    F. Alberto Grünbaum and Manuel D. de la Iglesia. Stochastic Darboux trans- formations for quasi-birth-and-death processes and urn models.Journal of Mathematical Analysis and Applications478(2019) 634–654.doi:10.1016/ j.jmaa.2019.05.048

    2019

  19. [26]

    de la Iglesia

    Manuel D. de la Iglesia. A note on the invariant distribution of a quasi-birth- and-death process.Journal of Physics A: Mathematical and Theoretical44 (2011) 135201.doi:10.1088/1751-8113/44/13/135201

    2011 doi

  20. [27]

    Mourad E. H. Ismail.Classical and Quantum Orthogonal Polynomials in One Variable. Vol. 98. Encyclopedia of Mathematics and its Applications. Cam- bridge University Press, 2009.doi:10.1017/CBO9781107325982

    2009 doi

  21. [28]

    Stanford: Stanford University Press, 1968

    Samuel Karlin.Total Positivity. Stanford: Stanford University Press, 1968

    1968

  22. [29]

    SamuelKarlinandJamesMcGregor.Randomwalks.Illinois Journal of Math- ematics3(1959) 66–81.doi:10.1215/ijm/1255454999

    1959

  23. [30]

    Samuel Karlin and James McGregor. The differential equations of birth-and- deathprocesses,andtheStieltjesmomentproblem.Transactions of the Amer- ican Mathematical Society85(1957) 489–546.doi:10.1090/S0002- 9947- 1957-0091566-1

    1957 doi

  24. [31]

    Prokhorov

    Jang Soo Kim, Abey López-García, and Valery A. Prokhorov. Matrix contin- ued fractions associated with lattice paths, resolvents of difference operators, and random polynomials.Constructive Approximation62(2025) 185–242. doi:10.1007/s00365-024-09685-1

    2025 doi

  25. [32]

    ASA-SIAM Series on Statistics and Applied Probability

    Guy Latouche and Vaidyanathan Ramaswami.Introduction to Matrix An- alytic Methods in Stochastic Modeling. ASA-SIAM Series on Statistics and Applied Probability. Philadelphia: Society for Industrial and Applied Math- ematics, 1999.doi:10.1137/1.9780898719734

    1999 doi

  26. [34]

    What is a multiple orthogonal polynomial?Notices of the AMS63(2016) 1029–1031.doi:10

    Andrei Martínez-Finkelshtein and Walter Van Assche. What is a multiple orthogonal polynomial?Notices of the AMS63(2016) 1029–1031.doi:10. 1090/noti1430. 96 REFERENCES

    2016

  27. [35]

    Nikishin and Vladimir N

    Evgenii M. Nikishin and Vladimir N. Sorokin.Rational Approximations and Orthogonality. Vol. 92. Translations of Mathematical Monographs. Provi- dence, RI: American Mathematical Society, 1991.doi:10.1090/mmono/092

    1991 doi

  28. [36]

    Allan Pinkus.Totally Positive Matrices. Vol. 181. Cambridge Tracts in Mathematics. Cambridge: Cambridge University Press, 2010.doi:10.1017/ CBO9780511691713

    2010

  29. [37]

    Cambridge: Cam- bridge University Press, 2008

    Lucy Joan Slater.Generalized Hypergeometric Functions. Cambridge: Cam- bridge University Press, 2008

    2008

  30. [38]

    Karlsson.Multiple Gaussian Hyperge- ometric Series

    Hari Mohan Srivastava and Per W. Karlsson.Multiple Gaussian Hyperge- ometric Series. Ellis Horwood Series in Mathematics and its Applications. Chichester: Ellis Horwood, 1985

    1985

  31. [39]

    Marcin J. Zygmunt. Matrix polynomials orthogonal with respect to a non- symmetric matrix of measures.Opuscula Mathematica36(2016) 409–423. doi:10.7494/OpMath.2016.36.3.409

    2016 doi

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.