Matrix Szeg\H{o} Function and Matrix Orthogonal Polynomials for Multiple Cuts

Sampad Lahiry

REVIEW 3 major objections 4 minor 25 references

Matrix Szeg\H{o} Function and Matrix Orthogonal Polynomials for Multiple Cuts

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper constructs the matrix Szegő factorisation for weights on several intervals and derives strong asymptotics for the associated matrix orthogonal polynomials.

desk verdict Fresh and largely right, but Proposition 3.4's pairing is non-holomorphic and needs a correction before the asymptotics can be trusted. read the letter →

arxiv 2608.11043 v1 pith:4A3PQW4C submitted 2026-08-11 math.CA

Sampad Lahiry This is my paper · ORCID

classification math.CA MSC 42C0530E2515A2314H6047B35

keywords matrix-valuedorthogonalpolynomialsmatrixSzegőfunctionmultiplecutsRiemann–HilbertproblemsteepestdescenthyperellipticRiemannsurfacevectorbundlesstrongasymptotics

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

The paper establishes a matrix analogue of the Szegő factorisation for matrix-valued weights supported on several disjoint intervals, and uses it to derive strong asymptotics for the associated matrix orthogonal polynomials. The central theorem says that if a positive definite matrix weight $M$ on a finite union of intervals is integrable enough, then $M=D_+D_+^*=D_-D_-^*$ almost everywhere, with $D$ holomorphic and invertible in the complement of the outermost interval; across each interior gap the boundary values differ by a constant unitary matrix, $D_+=D_-U_j$. That unitary gap jump is the new feature of the multi-cut setting. The paper then shows that two natural families of matrix weights—fixed Jacobi-type weights on the intervals and varying exponential weights whose equilibrium measure fills the intervals—both reduce under steepest descent to the same model Riemann–Hilbert problem, whose unique solution gives explicit asymptotic formulas for the matrix polynomials and their recurrence coefficients.

What carries the argument

The load-bearing object is the matrix Szegő function $D$, defined by the factorisation $M=D_\pm D_\pm^*$ with a unitary jump $D_+=D_-U_j$ on each gap; its construction rests on pulling the unit-circle matrix spectral factorisation back through a universal covering map, so that deck transformations are exactly the unitary jumps. For the asymptotic half, the load-bearing mechanism is the model Riemann–Hilbert problem whose jump matrices are the constant matrix $J$ on the cuts and $\mathrm{diag}(K_j,K_j^{-1})$ on the gaps, with $K_j=e^{-in\omega_j}U_j$. The paper solves this problem by viewing the rows of the parametrix as sections of a rank-$r$ flat unitary vector bundle $E_K$ over the hyperelliptic surface $y^2=A_0(z)B_0(z)$, with transition matrices $K_j$ across the gap curves. A bilinear pairing $\langle U,V\rangle_\tau$ built from the antiholomorphic involution $\tau$, together with a meromorphic differential $\eta=(A_0-B_0)/y\,dz$, yields a vanishing theorem stating that sections with poles only at the divisor $D_-$ and a zero at infinity do not exist; Riemann–Roch then upgrades the evaluation map at infinity to an isomorphism. The concrete parametrix ansatz with $c,s$ built from $\beta=\prod((z-b_\nu)/(z-a_\nu))^{1/4}$ converts the model problem into a pair of section problems. The uniqueness alternatives are a determinant argument showing $\det N\equiv 1$, and a vanishing-lemma plus Toeplitz-index computation in which the index is zero because the determinant curve $\det c^\#(\tau,\mu)=(\mu^2+(1-\mu)^2)^r$ stays in the positive real interval $[2^{-r},1]$.

What would settle it

Take $g=1$, choose a real analytic positive definite matrix weight $H$ on the two intervals (for instance a constant matrix), compute the actual monic matrix orthogonal polynomials numerically for $n=10,\ldots,100$, and compare $e^{-ng(z)}P_n(z)$ with the right-hand side of (6.17) uniformly on a compact set away from the cuts; if the difference does not decay like $O(n^{-1})$ as $n$ grows, the claimed local-parametrix transfer and hence the advertised asymptotic formula would be falsified.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is Theorem 1.4: for a positive definite matrix weight $M$ satisfying the assumed integrability conditions, there is a holomorphic, pointwise invertible matrix function $D$ on the complement of the outer interval, normalised by $D(\infty)>0$, with $M(x)=D_+(x)D_+(x)^*=D_-(x)D_-(x)^*$ for almost every $x$ in $E$, and with $D_+=D_-U_j$ on each gap, where $U_j$ is unitary. The factor $D$ is obtained by pulling the classical unit-circle matrix factorisation back through a universal covering map of the multi-cut domain; the deck transformations of that covering become the unitary jumps. This factorisation is then fed into a steepest descent analysis: after the $g$-function and lens transformations, both the Jacobi-type model and the varying exponential model reduce to one model problem with constant unitary jumps on the gaps, and the paper constructs its solution by encoding the unitary jump data in a flat unitary vector bundle over the hyperelliptic curve $y^2=A_0(z)B_0(z)$. Rows of the parametrix are sections of that bundle with a prescribed pole divisor $D_-$ and vanishing at one point at infinity, and an evaluation-at-infinity isomorphism provides the right number of sections. The resulting strong asymptotic formula is $e^{-ng(z)}P_n(z)=D(\infty)c(z)F_1(z;K_1,\ldots,K_g)D(z)^{-1}+O(n^{-1})$, uniformly on compact subsets of the complement of the outer interval.

Load-bearing premise

The load-bearing premise is that the local model solutions developed for a single interval can be transferred unchanged to each endpoint of the multi-cut problem, including internal endpoints where the extra unitary jump across the adjacent gap must be absorbed; the paper quotes this transfer rather than proving it in detail.

Editorial extensions

If this is right

For Jacobi-type matrix weights on several intervals, the monic matrix orthogonal polynomials satisfy $e^{-ng(z)}P_n(z)=D(\infty)c(z)F_1(z;K)D(z)^{-1}+O(n^{-1})$ uniformly on compact subsets of the complement of the outer interval; the same formula holds for varying exponential weights.
The gap data $K_j=e^{-in\omega_j}U_j$ are explicit: $U_j$ comes from the matrix Szegő function and $\omega_j$ from the equilibrium measure, so the effect of each gap on the polynomials is fully identified.
The model Riemann–Hilbert problem has exactly one solution for every choice of unitary matrices $K_j$, so the reduction from either polynomial model to the model problem is well-posed.
The vector-bundle construction also applies to scalar orthogonal polynomials: multi-cut scalar Szegő functions with modulus-one jumps across gaps can be read as meromorphic sections of flat unitary line bundles, giving a route that avoids explicit theta functions and special-divisor issues.
The asymptotic formulas imply corresponding asymptotics for the recurrence coefficients of the matrix polynomials, since those coefficients can be extracted from the expansion of the Riemann–Hilbert solution at infinity.

Reading between the lines

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

For the open direction the paper flags—weights of the form $W_n=W(x)^n$—the same machinery would need the $g$-function analysis on the spectral curve of the eigenvalues rather than on the fixed hyperelliptic curve; formula (6.28) gives a concrete target to test there.
The unitary-jump structure of the matrix Szegő function suggests that multi-cut matrix factorisation could be phrased more generally as a flat unitary connection on the complement of $E$, with the $U_j$ as monodromy matrices; the paper does not develop this interpretation.
A numerical check of the leading term in (6.17) for $g=1$, with a constant matrix weight $H$ and $U_1$ determined by the Szegő factor, would isolate the effect of the gap jump from the local endpoint analysis and could reveal how quickly the $O(n^{-1})$ error sets in.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: this is a real contribution with a genuine flaw in the main proof. Theorem 1.4, the matrix Szegő factorization for multiple intervals with unitary jumps on gaps, is new and, as far as I can tell, correct; the proof via Wiener–Masani and deck transformations is clean. The vector-bundle construction of the global parametrix is also a good idea and gives the right picture. The problem is Proposition 3.4. The pairing U(P)V(τP)^T is not holomorphic: τ is antiholomorphic, so V(τP) is antiholomorphic in general. The proof conflates transpose with conjugate transpose—it uses K_jK_j^T=K_jK_j^*=I and states that on ∂S the pairing equals UV^*, which only holds for U V^*. Theorem 3.10, Proposition 3.12, Lemma 4.4, and Proposition 4.5 all depend on this, so the existence proof for RH Problem 1.10 as written does not go through. Section 5 gives a separate Fredholm proof that looks correct, but Section 6 uses the bundle solution, so the final asymptotic formulas inherit the gap.

The good news is that the error looks fixable. Replace V(τP)^T with V(τP)^*. Then the pairing becomes holomorphic in the appropriate sense, the jump across Γ_j matches by unitarity, and on ∂S it becomes UU^*, so the Stokes argument in Theorem 3.10 goes through. I checked the divisor bookkeeping; it works. This is not a dead end, but it is a load-bearing correction, not a typo in a remark.

The reader's other concern is fair: the transfer of the local Bessel and Airy parametrices from the single-cut papers to internal endpoints is asserted rather than proved. The conjugation by e^{-inω_jσ3} plausibly removes the extra gap jump, but the paper should show the computation. The endpoint matching and N4 bounds need a real check, not a citation. That said, the overall strategy is standard enough that I expect it to work.

The paper is for specialists in Riemann–Hilbert steepest descent and matrix orthogonal polynomials. It deserves a serious referee—don't desk reject it—but I would not accept it without the Proposition 3.4 correction and a detailed check of the local-parametrix transfer. If those come back clean, the paper is likely correct and important.

Referee Report

3 major / 4 minor

Summary. The paper constructs a matrix Szegő factorization for matrix-valued weights supported on a finite union of intervals, showing that the factor has constant unitary jumps across the gaps. It then builds the global parametrix for the associated Riemann–Hilbert problem using sections of flat unitary vector bundles over the hyperelliptic curve determined by the multi-cut set, and applies this machinery to obtain strong asymptotics for matrix-valued orthogonal polynomials in two settings: fixed Jacobi-type weights on multiple intervals and varying exponential weights whose equilibrium measure has multi-component support. An alternative proof of solvability via Toeplitz-index theory is also presented in Section 5.

Significance. If the results are correct, they would constitute a substantial extension of the one-cut asymptotics of Deaño–Kuijlaars–Román to multi-interval supports, and the vector-bundle viewpoint would offer a new route to global parametrices that avoids theta functions and special divisors. The paper is well structured and contains several useful auxiliary lemmas, and the Fredholm-index alternative in Section 5 is of independent interest. However, the main existence proof for the global parametrix currently rests on an invalid holomorphy claim in Proposition 3.4, so the central results are not yet established as they stand.

major comments (3)

[Section 3, Proposition 3.4] The pairing ⟨U,V⟩_τ(P) = U(P)V(τP)^T is not meromorphic on X, because τ is anti-holomorphic and therefore V(τP) is anti-holomorphic in P; for example, in a local holomorphic coordinate z away from ∂S, V(z)=z gives V(τP)=z̄, which is not holomorphic. Consequently the form ω = ⟨U,U⟩_τ η in the proof of Theorem 3.10 is not a holomorphic 1-form, and the argument dω=0 with Stokes' theorem has no basis. Moreover, on ∂S the claimed identity ⟨U,V⟩_τ = UV^* is false with the transpose; the correct identity requires the conjugate transpose. The likely fix is to define ⟨U,V⟩_τ(P) = U(P)\overline{V(τP)}^T, which is holomorphic and satisfies the desired boundary identity. As stated, Theorem 3.10, Proposition 3.12, Lemma 4.4, and Proposition 4.5 are not proved, and the existence of the solution to RH Problem 1.10 used in Section 6 is not established. This is a load-bearing defect that must be repaired.
[Section 5, Subsections 5.2.1–5.2.3] The Fredholm route proves invertibility of the operator T on L^2, yielding a solution of the form N = I + Cϕ with L^2 boundary values. However, it is not shown that this L^2 solution satisfies the pointwise endpoint condition (N4) of RH Problem 1.10, namely N(z) = O(|z-e|^{-1/4}) at each endpoint. Without such an estimate, Section 5 cannot replace the bundle construction in the applications of Section 6, where the endpoint behavior is needed for the matching step.
[Section 6, Subsections 6.1.4 and 6.2.4] The local Bessel/Airy parametrices of [6, §3.5] and [7, §5.5] are transferred to the multi-cut setting by assertion rather than by proof. In particular, the absorption of the additional unitary jump e^{-inω_j σ3} on the gap adjacent to an internal endpoint into the prefactor E_{e,n} is described only in a few sentences. Since the local matching is essential for the asymptotic formulas (6.17) and (6.28), the author should either provide the full details of this transfer or give a precise reference where it is carried out for multi-cut problems.

minor comments (4)

[Section 3, Proposition 3.4] There is a stray period and semicolon in the sentence 'The isolated branch points are removable because the function is locally bounded there whenever U,V are holomorphic there; .'
[Section 3, Lemma 3.5] The phrase 'has a simple zero at ξ_j' is missing the word 'at', and the final clause 'these are all the g zeros' should be rephrased for clarity.
[Section 6.2.4] The phrase 'Following [7, (5.24)] and 1.8' should presumably read 'and Assumption 1.8' rather than 'and 1.8'.
[Section 6, Proposition 6.5] The block-matrix computation for D^{-1}_- and D_+ is dense; an explicit display of D_± in terms of D_± and D_±^* would improve readability and help verify the jump calculations.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the matrix Szegő construction and the RH 1.10 parametrix are built from external inputs and explicit transformations, not from the target asymptotics.

full rationale

The derivation chain is not circular. Theorem 1.4 obtains D from the Wiener–Masani factorization via a universal covering map, with the unitary gap matrices U_j arising from deck transformations of the covering, not from the later asymptotic formulas. The global parametrix for RH Problem 1.10 is constructed from sections of the flat unitary bundle E_K with the data K_j fixed in Definition 3.3; existence rests on the evaluation-map isomorphism of Proposition 3.12, which combines Theorem 3.10 with Riemann–Roch, and uniqueness on detN≡1 and Liouville’s theorem. Section 5 supplies an independent Fredholm/Toeplitz route using the external index theorem of [3]. Section 6 reduces each model to RH Problem 1.10 by explicit steepest-descent transformations and defines K_j = e^{-inω_j}U_j, so no fitted parameter is renamed as a prediction. The skeptical concerns are real mathematical-validity issues rather than circularity: Proposition 3.4 asserts that P ↦ V(τP) is holomorphic although τ is anti-holomorphic, and Sections 6.1.4/6.2.4 transfer the single-interval local parametrices of [6] and [7] to internal endpoints without proof. Neither is a reduction of a conclusion to its own inputs, so the circularity score remains 0.

Assumptions & free parameters 0 free parameters · 8 assumptions · 1 invented entities

The results depend on standard theorems: Wiener-Masani factorization, uniformization and radial limit theorems, Riemann-Roch, Toeplitz index theory, and equilibrium measure theory. No parameters are fitted. The only new object, the bundle E_K, is explicitly constructed from the jump data, so the circularity burden is minimal.

assumptions (8)

standard math Wiener-Masani spectral factorization theorem (Theorem 1.1, [24])
Provides the outer H^2 factor G0 for the pulled-back weight on the unit circle; used in the proof of Theorem 1.4.
standard math Uniformization theorem and universal covering map p:D→Ω with p(0)=∞
Used to pull the multi-cut domain back to the disk; invoked in Section 1.1 around (1.8).
standard math Radial limits of universal covering maps (Ferreira-Jové, [12, Cor. B])
Gives boundary values p_*(τ) for almost every τ, needed to define D_± from G0.
standard math Riemann-Roch theorem for holomorphic vector bundles (Theorem 3.11, [15])
Used to prove h0(V±)=r and the surjectivity of evaluation at infinity in Proposition 3.12.
standard math Toeplitz operator Fredholm and index theorem for piecewise continuous symbols (Theorem 5.5, [3])
Used to prove the singular integral operator T has index 0 in Proposition 5.6.
standard math Equilibrium measure and harmonic measure properties for compact sets and external fields (Saff-Totik [21]; Deift-Kriecherbauer-McLaughlin-Venakides-Zhou [9])
Used for g-function identities and the strictly negative real part of φ in Sections 6.1 and 6.2.
standard math Scalar outer function theory on the disk (Nikolski [20])
Used for the scalar factor h in Lemma 6.1 and for outer factor properties in Lemma 1.2.
domain assumption Local Bessel (Deaño-Kuijlaars-Román [6]) and Airy (Deaño-Román [7]) parametrices transfer unchanged to each endpoint
Sections 6.1.4 and 6.2.4 quote these local solutions rather than reproving them; this is necessary for the strong asymptotics formulas.

invented entities (1)

Flat unitary rank-r vector bundle E_K over the hyperelliptic surface X
purpose: Encodes the unitary jump matrices K_j on the gaps so that the global parametrix can be realized as holomorphic sections with prescribed poles at the divisor D±.
Mathematical construction defined in Definition 3.3 from the given data K_j; its consistency is verified by the cocycle condition. It is not an empirical entity and requires no external falsifiable evidence.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Matrix Szeg\H{o} Function and Matrix Orthogonal Polynomials for Multiple Cuts." pith.science (2026). https://pith.science/paper/4A3PQW4C

@misc{pith2026260811043,
  author       = {Pith},
  title        = {Pith review of: Matrix Szeg\Ho Function and Matrix Orthogonal Polynomials for Multiple Cuts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4A3PQW4C}},
  note         = {Machine review of arXiv:2608.11043}
}

read the original abstract

Following recent developments on matrix-valued orthogonal polynomials (MVOPs), we construct the matrix Szeg\H{o} factorisation of the matrix weight function when it is supported on multiple intervals. We find that on the gaps the matrix function has unitary multiplicative jumps. In the next part, performing the Deift-Zhou steepest descent analysis we study the associated global parametrix which comes from the Riemann-Hilbert problem of the MVOPs. A solution is constructed visualising the rows of the parametrix as sections of vector bundles on a hyper-elliptic Riemann surface. An alternate view point to the global parametrix is also presented using a vanishing lemma. As an application we obtain strong asymptotics of MVOPs for multiple cuts which extends the works of Dea\~no, Kuijlaars, and Rom\'an.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages

[1]

A. I. Aptekarev,Asymptotic properties of polynomials orthogonal on a system of contours, and periodic motions of Toda lattices, Mat. Sb. (N.S.)125(167)(1984), 231–258; English transl. Math. USSR Sb.53(1986), 233–260

work page 1984
[2]

Bleher and K

P. Bleher and K. Liechty,Random Matrices and the Six-Vertex Model, CRM Monograph Series, vol. 32, American Mathematical Society, Providence, RI, 2014

work page 2014
[3]

Böttcher and B

A. Böttcher and B. Silbermann,Analysis of Toeplitz Operators, 2nd ed., Springer Mono- graphs in Mathematics, Springer, Berlin, 2006. 33

work page 2006
[4]

G. A. Cassatella-Contra and M. Mañas,Riemann–Hilbert problems, matrix orthogonal polynomials and discrete matrix equations with singularity confinement, Stud. Appl. Math. 128(2012), no. 3, 252–274

work page 2012
[5]

Damanik, A

D. Damanik, A. Pushnitski, and B. Simon,The analytic theory of matrix orthogonal polynomials, Surv. Approx. Theory4(2008), 1–85

work page 2008
[6]

Deaño, A

A. Deaño, A. B. J. Kuijlaars, and P. Román,Asymptotics of matrix valued orthogonal polynomials on[−1,1], Adv. Math.423(2023), 109043

work page 2023
[7]

Asymptotics of matrix orthogonal polynomials on the real line

A. Deaño and P. Román,Asymptotics of matrix orthogonal polynomials on the real line, arXiv:2508.04908, 2025

work page Pith review arXiv 2025
[8]

Deift,Orthogonal Polynomials and Random Matrices: A Riemann–Hilbert Approach, Courant Lecture Notes in Mathematics, vol

P. Deift,Orthogonal Polynomials and Random Matrices: A Riemann–Hilbert Approach, Courant Lecture Notes in Mathematics, vol. 3, American Mathematical Society, Provi- dence, RI, 1999

work page 1999

Show all 25 references

[9]

Deift, T

P. Deift, T. Kriecherbauer, K. T.-R. McLaughlin, S. Venakides, and X. Zhou,Uniform asymptotics for polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory, Comm. Pure Appl. Math. 52(1999), 1335–1425

1999
[10]

Deift, T

P. Deift, T. Kriecherbauer, K. T.-R. McLaughlin, S. Venakides and X. Zhou, A Riemann– Hilbert approach to asymptotic questions for orthogonal polynomials, J. Comput. Appl. Math.133(2001), 47–63

2001
[11]

Ephremidze, G

L. Ephremidze, G. Janashia, and E. Lagvilava,An analytic proof of the matrix spectral factorization theorem, Georgian Math. J.15(2008), no. 2, 241–249

2008
[12]

G. R. Ferreira and A. Jové,Boundary behaviour of universal covering maps, Adv. Math. 469(2025), Article 110232

2025
[13]

A. S. Fokas, A. R. Its, and A. V. Kitaev,The isomonodromy approach to matrix models in 2D quantum gravity, Comm. Math. Phys.147(1992), 395–430

1992
[14]

F. A. Grünbaum, M. D. de la Iglesia, and A. Martínez-Finkelshtein,Properties of matrix orthogonal polynomials via their Riemann–Hilbert characterization, SIGMA Symmetry Integrability Geom. Methods Appl.7(2011), Paper 098, 31 pp

2011
[15]

R. C. Gunning,Lectures on Vector Bundles over Riemann Surfaces, Princeton Mathe- matical Notes, No. 6, Princeton University Press, Princeton, NJ, 1967

1967
[16]

A. B. J. Kuijlaars, Matrix valued orthogonal polynomials arising from hexagon tilings with3×3-periodic weightings, J. Approx. Theory311(2025), Article 106202

2025
[17]

A. B. J. Kuijlaars and K. T.-R. McLaughlin,Generic behavior of the density of states in random matrix theory and equilibrium problems in the presence of real analytic external fields, Comm. Pure Appl. Math.53(2000), no. 6, 736–785

2000
[18]

A. B. J. Kuijlaars, K. T.-R. McLaughlin, W. Van Assche, and M. Vanlessen,The Riemann– Hilbert approach to strong asymptotics for orthogonal polynomials on[−1, 1], Adv. Math. 188(2004), no. 2, 337–398

2004
[19]

A. B. J. Kuijlaars and M. Vanlessen,Universality for eigenvalue correlations at the origin of the spectrum, Comm. Math. Phys.243(2003), no. 1, 163–191. 34

2003
[20]

Nikolski,Hardy Spaces, translated by D

N. Nikolski,Hardy Spaces, translated by D. Gibbons and G. Gibbons, Cambridge Studies in Advanced Mathematics, vol. 179, Cambridge University Press, Cambridge, 2019

2019
[21]

E. B. Saff and V. Totik,Logarithmic Potentials with External Fields, Grundlehren der mathematischen Wissenschaften, vol. 316, Springer-Verlag, Berlin, 1997

1997
[22]

Volberg and P

A. Volberg and P. Yuditskii,Kotani–Last problem and Hardy spaces on surfaces of Widom type, Invent. Math.197(2014), no. 3, 683–740

2014
[23]

Widom,Extremal polynomials associated with a system of curves in the complex plane, Advances in Math.3(1969), 127–232

H. Widom,Extremal polynomials associated with a system of curves in the complex plane, Advances in Math.3(1969), 127–232

1969
[24]

Wiener and P

N. Wiener and P. Masani,The prediction theory of multivariate stochastic processes. I. The regularity condition, Acta Math.98(1957), 111–150

1957
[25]

Zhou,The Riemann–Hilbert problem and inverse scattering, SIAM J

X. Zhou,The Riemann–Hilbert problem and inverse scattering, SIAM J. Math. Anal.20 (1989), no. 4, 966–986. 35

1989

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

A. I. Aptekarev,Asymptotic properties of polynomials orthogonal on a system of contours, and periodic motions of Toda lattices, Mat. Sb. (N.S.)125(167)(1984), 231–258; English transl. Math. USSR Sb.53(1986), 233–260

work page 1984

[2] [2]

Bleher and K

P. Bleher and K. Liechty,Random Matrices and the Six-Vertex Model, CRM Monograph Series, vol. 32, American Mathematical Society, Providence, RI, 2014

work page 2014

[3] [3]

Böttcher and B

A. Böttcher and B. Silbermann,Analysis of Toeplitz Operators, 2nd ed., Springer Mono- graphs in Mathematics, Springer, Berlin, 2006. 33

work page 2006

[4] [4]

G. A. Cassatella-Contra and M. Mañas,Riemann–Hilbert problems, matrix orthogonal polynomials and discrete matrix equations with singularity confinement, Stud. Appl. Math. 128(2012), no. 3, 252–274

work page 2012

[5] [5]

Damanik, A

D. Damanik, A. Pushnitski, and B. Simon,The analytic theory of matrix orthogonal polynomials, Surv. Approx. Theory4(2008), 1–85

work page 2008

[6] [6]

Deaño, A

A. Deaño, A. B. J. Kuijlaars, and P. Román,Asymptotics of matrix valued orthogonal polynomials on[−1,1], Adv. Math.423(2023), 109043

work page 2023

[7] [7]

Asymptotics of matrix orthogonal polynomials on the real line

A. Deaño and P. Román,Asymptotics of matrix orthogonal polynomials on the real line, arXiv:2508.04908, 2025

work page Pith review arXiv 2025

[8] [8]

Deift,Orthogonal Polynomials and Random Matrices: A Riemann–Hilbert Approach, Courant Lecture Notes in Mathematics, vol

P. Deift,Orthogonal Polynomials and Random Matrices: A Riemann–Hilbert Approach, Courant Lecture Notes in Mathematics, vol. 3, American Mathematical Society, Provi- dence, RI, 1999

work page 1999