Recurrence Coefficients of the Orthogonal Polynomials for Oscillatory Jacobi-type Weight Functions

Shulin Lyu; Xun Zhou

REVIEW 2 major objections 5 minor 25 references

Recurrence Coefficients of the Orthogonal Polynomials for Oscillatory Jacobi-type Weight Functions

T0 review · 2 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper derives two coupled difference equations that determine all three-term recurrence coefficients for monic orthogonal polynomials with oscillatory Jacobi-type weights w1(x)=(1−x^2)^{λ−1/2}e^{iζx} and w2(x)=x(1−x^2)^{γ−1/2}e^{iηx} o

desk verdict Solid, useful difference-equation work with one load-bearing yet fixable gap: the ladder-operator framework is quoted under positivity and extended to complex weights without proof; plus a missing conclusions section. read the letter →

arxiv 2607.19797 v1 pith:VLNFAVII submitted 2026-07-22 math-ph math.MP

Shulin Lyu , Xun Zhou This is my paper · ORCID

classification math-phmath.MP MSC 33C4542C0533C10

keywords oscillatoryGegenbauerweightJacobi-typethree-termrecurrencecoefficientsladderoperatorscompatibilityconditionsdifferenceequationsorthogonalpolynomialsBesselfunctions

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 sets out to show that the three-term recurrence coefficients of orthogonal polynomials for two oscillatory Jacobi-type weights are governed by simple coupled difference equations. For the oscillatory Gegenbauer weight (1−x^2)^{λ−1/2}e^{iζx}, it proves equations (2.3)–(2.4) that need only four initial values; for the weight x(1−x^2)^{γ−1/2}e^{iηx}, it proves equations (3.19)–(3.20). If these hold, all coefficients can be generated iteratively, and the equations are structurally simpler and lower-order than previously known ones. The paper also uses the generated tables to conjecture exact rational symbolic forms for α_n and β_n, offering an explicit road from the weight to the recurrence data needed for Gaussian quadrature of oscillatory integrals.

What carries the argument

The central object is the pair of three-term recurrence coefficients α_n, β_n, together with the ladder operators (1.11)–(1.12) and the three compatibility conditions (S1), (S2), and (S′2) from Proposition 1.1. These conditions relate the raising-and-lowering functions A_n(z) and B_n(z) through identities that must hold for every n; substituting explicit rational expressions for A_n and B_n computed from the oscillatory weights and eliminating auxiliary quantities (p(n), R_n, r_n) yields the coupled difference equations. A second load-bearing ingredient is the Bessel-function condition on ζ and η — zeros of J_{λ−1} or J_0, or η=mπ — which guarantees existence of the orthogonal polynomials an

What would settle it

Take λ=0 and ζ the first positive zero of J_{−1}; compute α_4 and β_4 directly from the inner products h_n using high-precision numerical quadrature of (1−x^2)^{−1/2}e^{iζx}, and compare with the values from iterating (2.3)–(2.4) from (2.5). A difference beyond integration round-off would refute Theorem 2.2. Similarly, take γ=0 and η a positive zero of J_0; direct quadrature of x(1−x^2)^{−1/2}e^{iηx} against iteration of (3.19)–(3.20) would decide Theorem 3.4.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.2 for w1 and Theorem 3.4 for w2: the monic orthogonal polynomials' recurrence coefficients satisfy coupled difference equations, so once a small set of initial coefficients is fixed, all later coefficients are determined. The derivations substitute explicit formulas for the ladder quantities A_n(z) and B_n(z) into the three compatibility conditions and eliminate auxiliary integrals; the initial values themselves come from known Bessel-function integral identities. The paper further observes that the iterated coefficients are real for w1 and rational in the special w2 cases, and formulates Conjectures 2.5–3.9 giving closed rational expressions in ζ^2 or η^2 with

Load-bearing premise

The whole derivation rests on the assertion that the ladder operators and compatibility conditions, originally proved for a positive base weight w0(x)>0, also hold for the complex-valued oscillatory factors e^{iζx} and x e^{iηx}; the paper states this follows because the positivity condition was not used in the original proof, but does not give a proof of that extension.

Editorial extensions

If this is right

For w1, the four numbers α0, α1, β1, and β2 determine every α_n and β_n through (2.3)–(2.4), independently of β0.
For w2, starting from α0,...,α3 and β1,...,β4, equations (3.19)–(3.20) generate all later coefficients.
The equations are structurally simpler than existing ones: for w1, equation (2.3) is first-order in β, replacing a second-order equation in prior literature.
Iterating reproduces known numerical tables, including the γ=1/2, η=100π case previously computed by Gaussian quadrature methods, confirming internal consistency.
The pattern of computed coefficients motivates explicit rational symbolic conjectures for all α_n and β_n in terms of monic polynomials in ζ^2 or η^2.

Reading between the lines

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

The same ladder-operator route may apply to other oscillatory Jacobi-type weights whose log-derivative is a rational function of x, such as (1−x)^α(1+x)^β e^{iζx} with α≠β; a direct test would be to derive analogous difference equations in that case.
The conjectured symbolic forms, with monic polynomials indexed by k^2 and k(k±1), suggest these polynomials are Hankel determinants or related combinatorial objects; one testable route is to verify they satisfy a finite Toda-type recurrence in k.
If the conjectures hold, the explicit rational expressions could be expanded to give direct large-n asymptotics for α_n and β_n without solving the difference equations numerically.

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.

Desk Editor's Note

read the letter

The paper does something genuinely new. For the oscillatory Gegenbauer weight w1 it reduces the known second-order difference equation to a first-order one (2.3), and for w2 = x(1-x^2)^{γ-1/2} e^{iηx} it gives the first derived difference equations for the recurrence coefficients. The symbolic conjectures are new and clearly labelled as conjectures. The derivations are mostly transparent, the initial values are computed in detail (including hand derivations in appendices), and the tables agree with earlier results where they overlap. The Mathematica code and the numerical check against [14] give real evidence that the equations are correct.

The soft spot is the foundation. Proposition 1.1 is quoted from [12] under w0(x)>0. The paper asserts in one sentence that positivity was not used in the derivation and therefore the proposition applies to the complex weights (1.1)-(1.2). No proof is offered. This matters most for w2, where w0 = x e^{iηx} is sign-changing and v'(x) has a simple pole at 0. The A_n, B_n expressions in Lemma 3.1 formally cancel the pole in the difference quotient, and the numerical checks pass, so the gap is probably fixable rather than fatal. But the central theorems inherit an unproven hypothesis, and a referee should ask for either a direct proof of the ladder operators and compatibility conditions for these weights or a clear reference where the extension is established. Theorem 3.4 also divides by β_n−β_{n−1} and α_{n−2}−α_{n−1} without stating that they are nonzero; the special cases in the tables have nonzero denominators, but the theorem as stated needs a nonvanishing assumption or a separate argument. Those are the two substantive issues.

Minor: the introduction promises conclusions in Section 4, and no Section 4 exists. That is an editorial slip, easily fixed.

Bottom line: the algebra and the numerical evidence are strong enough that I would not desk-reject this. It deserves a serious referee, with the ladder-operator extension and denominator issues flagged. For someone working on oscillatory weights or Gaussian quadrature, the paper is useful and citeable. I would probably bring it to a reading group, mostly to see how the ladder-operator gap gets resolved.

Referee Report

2 major / 5 minor

Summary. The manuscript studies two classes of oscillatory Jacobi-type weights: w1(x)=(1-x^2)^{λ-1/2}e^{iζx} and w2(x)=x(1-x^2)^{γ-1/2}e^{iηx}. For each, the authors compute the ladder-operator quantities A_n(z), B_n(z) explicitly in terms of the recurrence coefficients {α_n, β_n} and auxiliary quantities, then apply the compatibility conditions (S1), (S2), (S'_2) to derive coupled difference equations for α_n and β_n. The main results are Theorem 2.2, giving (2.3)-(2.4) for w1 with explicit initial values (2.5), and Theorem 3.4, giving (3.19)-(3.20) for w2 after eliminating p(n) from the system (3.16)-(3.18). Tables 1-4 list computed coefficients for special cases, and several conjectured symbolic forms are proposed in Conjectures 2.5, 2.7, 3.7, and 3.9. The paper claims that these difference equations are structurally simpler and of lower order than existing ones and that, once initial values are known, the recurrence coefficients can be iteratively computed.

Significance. If the central derivations are valid, the paper delivers explicit, low-order coupled difference equations for the recurrence coefficients of two nontrivial non-Hermitian orthogonal-polynomial systems. This would be a useful improvement over the existing literature: for w1 the equations are indeed simpler than those in [15], and for w2 the paper provides explicit iterable equations not previously written down. The extensive tables, appendices with hand derivations of initial values, and Mathematica code are valuable and make the algebraic steps reproducible. The conjectures are clearly labeled as conjectures and do not affect the correctness of the difference equations. However, the entire derivation rests on the unproved extension of Proposition 1.1 from positive w0 to complex-valued, sign-changing w0, and the final difference equations contain divisions by quantities whose nonvanishing is not established. These are load-bearing issues, so the central claim is only conditionally established.

major comments (2)

[Section 1, Proposition 1.1] The manuscript's central tool is Proposition 1.1, restated from [12] under the hypothesis w0(x)>0. The sentence 'the condition w0(x)>0 ... was not used in the derivation' is an assertion, not a proof. This matters especially for w2(x)=x(1-x^2)^{γ-1/2}e^{iηx}: here w0(x)=x e^{iηx} is not positive and, as shown in (3.2), v'(x) has a simple pole at x=0. The formal cancellation of that pole in the difference quotient leading to Lemma 3.1 does not by itself justify the integration-by-parts identities and boundary-term structure used in [12]. Since Lemmas 2.1 and 3.1 and hence Theorems 2.2 and 3.4 are algebraic consequences of Proposition 1.1, the unproved extension is load-bearing. The authors should either prove that the ladder-operator theorem holds for the complex, sign-changing weights (1.1)-(1.2), or give a precise reference where it is proved under hypotheses covering these cases.
[Theorem 3.4, Eqs. (3.19)-(3.20)] The proof of Theorem 3.4 solves (3.18) and (3.17) for p(n) and p(n-1), respectively. This introduces explicit divisions by α_{n-1}-α_n and β_n-β_{n-1}; moreover both (3.19) and (3.20) contain β_{n+1} in the denominator, and (3.20) also contains β_{n-1}-β_n in the denominator. The paper nowhere proves that these factors are nonzero in the parameter regimes considered. Consequently the iterative claim in the paragraph after (3.19)-(3.20) is only formal unless a genericity assumption is added. The same remark applies to Theorem 2.2, where (2.3) contains β_{n+1} in a denominator, and to the initial-value formulas (2.5), which require ζ^2≠2λ. If nonvanishing follows from known existence theorems or from the explicit forms in Tables 1-4, that argument should be supplied; otherwise the statements should be qualified.

minor comments (5)

[Introduction, final paragraph] The text says 'Our conclusions are presented in Section 4,' but the manuscript ends with Section 3 and appendices. Either add a conclusions section or correct the cross-reference.
[Section 3, paragraph after (3.18)] The sentence 'When λ=0 and ζ is a positive zero of the Bessel function J0, or λ=1/2 and ζ=mπ' uses the wrong symbols; it should read 'γ=0 and η...' and 'γ=1/2 and η=mπ'.
[Lemma 3.1, Eqs. (3.3)-(3.4)] The auxiliary quantities R_n(ζ) and r_n(ζ) are defined in (3.5)-(3.6) with exp(iηx), so the argument should be η, not ζ. The notation is inconsistent.
[Throughout] Typographical errors: 'Corollay' in Corollary 2.4, 3.5, and 3.8; 'Frourier-type' in reference [9]; 'G V. Milovanović' in reference [14] should be 'G. V. Milovanović'.
[Remark 3.12] The remark reports last-digit discrepancies with [14, Table 2] without comment. A brief explanation of rounding precision would be helpful, since a reader may otherwise wonder whether the difference equations or the code are at fault.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reduction found: the recurrence-coefficient equations are obtained from ladder-operator identities and explicit moment integrals, not from the conjectures or from the target equations themselves.

full rationale

The derivation chain is not circular by construction. The paper starts from explicit expressions for A_n(z) and B_n(z) obtained by substituting v'(x) into the definitions (1.13)-(1.14), then substitutes those expressions into the compatibility conditions (S_1), (S_2) and (S'_2) and eliminates auxiliary quantities. For w_1, Lemma 2.1 and the algebra in the proof of Theorem 2.2 directly produce (2.3)-(2.4). For w_2, Lemma 3.1, Proposition 3.3 and the elimination in Theorem 3.4 similarly produce (3.19)-(3.20). Initial values are computed independently in Appendices A and E from the moment integrals, using Bessel-function identities, not from the target equations. Thus no 'predicted' α_n or β_n is a fitted parameter renamed as a prediction, and no equation used in the derivation is equivalent to its own conclusion. The conjectures in Sections 2 and 3 are explicitly pattern guesses from the iterated tables, and they are not used to derive any theorem. The main caveat is not circularity: Proposition 1.1 is quoted from [12], co-authored by the first author, and the paper asserts without proof that 'the condition w0(x)>0 is imposed to guarantee the existence of the orthogonal polynomials, and it was not used in the derivation... Therefore, Theorem 4.1 of [12] also applies' to the complex weights (1.1)-(1.2). This unproved extension is a load-bearing rigor/correctness risk, but it does not reduce the derived difference equations to their inputs — the ladder-operator theorem is an external framework, and the paper's own numerical/Mathematica results are separately checked where possible against [14] and [15]. Editorial defects such as the promised Section 4 being absent and some tables being hard to parse are also not circularity. Overall, no specific circular step can be exhibited, so the circularity score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claims rest on four premises: the ladder-operator framework for complex weights (taken from a same-author paper), the existence of the orthogonal polynomial sequences from prior literature, and a non-vanishing denominator condition for the second family. There are no free parameters: the initial coefficients are derived from Bessel-function integrals, and the conjectured polynomials are outputs, not inputs.

assumptions (4)

domain assumption Ladder operators (1.11)-(1.12) and compatibility conditions (S1),(S2),(S'2) hold for the complex, non-positive weights (1.1) and (1.2).
Proposition 1.1 restates Theorem 4.1 of [12], where a positivity condition w0(x)>0 was part of the hypotheses. The paper asserts that positivity was not used, but gives no proof here; since [12] shares the first author, this is not independently verified.
domain assumption For w1(x)=(1−x^2)^{λ−1/2}e^{iζx} with rational λ>−1/2 and ζ a real nonzero zero of J_{λ−1}, a unique sequence of monic orthogonal polynomials exists.
Invoked from [15, Theorem 2.2] (extending [16]); used to justify h_n ≠ 0 and the recurrence (1.5).
domain assumption For w2(x)=x(1−x^2)^{γ−1/2}e^{iηx}, monic orthogonal polynomials exist when γ=0 and η is a positive zero of J_0, or when γ=1/2 and η=mπ for a nonzero integer m.
Invoked from [13, Theorem 4] and [14, Theorem 2.3]; used in Corollaries 3.5 and 3.8.
ad hoc to paper The differences β_n−β_{n−1} and α_{n−2}−α_{n−1} are nonzero for all n≥3 in the parameter regimes considered.
Equations (3.19)-(3.20) are derived by dividing by these differences; the paper does not state or prove the non-vanishing.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Recurrence Coefficients of the Orthogonal Polynomials for Oscillatory Jacobi-type Weight Functions." pith.science (2026). https://pith.science/paper/VLNFAVII

@misc{pith2026260719797,
  author       = {Pith},
  title        = {Pith review of: Recurrence Coefficients of the Orthogonal Polynomials for Oscillatory Jacobi-type Weight Functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VLNFAVII}},
  note         = {Machine review of arXiv:2607.19797}
}

abstract

We study two classes of oscillatory Jacobi-type weight functions $x^c(1-x^2)^{\lambda-1/2}\exp(i\zeta x)$, $x\in[-1,1], \lambda>-1/2, c\in\{0,1\}$. Other restrictions are imposed on $\lambda$ and $\zeta$ to guarantee the existence of the associated orthogonal polynomials. By using the ladder operators established in the recent literature for monic orthogonal polynomials associated with Jacobi-type weight functions and three compatibility conditions, we derive two coupled difference equations satisfied by the three-term recurrence coefficients. Compared with the existing results, these equations are structurally simpler and of lower order. Once the initial values are determined, the recurrence coefficients can be computed at any stage through the difference equations. The obtained expressions enable us to conjecture the symbolic forms for the recurrence coefficients.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 1 linked inside Pith

[14]

Milovanovi´ c and A

G V. Milovanovi´ c and A. S. Cvetkovi´ c, Orthogonal polynomials and Gaussian quadrature rules related to oscillatory weight functions, J. Comput. Appl. Math. 179 (2005), 263-287

2005
[12]

Lyu and Y

S. Lyu and Y. Lyu, Ladder Operators for Laguerre-type and Jacobi-type Orthogonal Poly- nomials, J. Phys. A: Math. Theor.59(2026), 115201 (30pp)

2026
[15]

G. V. Milovanovi´ c, A. S. Cvetkovi´ c and Z. M. Marjanovic, Orthogonal polynomials for the oscillatory-Gegenbauer weight, Publ. Inst. Math. (Belgr.) (N.S.)84(2008), 49-60

2008
[1]

Basor, Y

E. Basor, Y. Chen and N. Haq, Asymptotics of determinants of Hankel matrices via non- linear difference equations, J. Approx. Theory198(2015), 63-110

2015
[2]

Chen and M

Y. Chen and M. Ismail, Jacobi polynomials from compatibility conditions, Proc. Am. Math. Soc.133(2005), 465-472

2005
[3]

Chen and A

Y. Chen and A. Its, Painlev´ e III and a singular linear statistics in Hermitian random matrix ensembles, I, J. Approx. Theory162(2010), 270-297

2010
[4]

Chen and M

Y. Chen and M. McKay, Coulomb fluid, Painlev´ e transcendents and the information theory of MIMO systems, IEEE Trans. Inf. Theory58(2012), 4594-4634

2012
[5]

R. Chen, D. Yu and J. Chen, Computation of oscillatory integrals with an exponential kernel and Jacobi-type singularities, J. Math. Anal. Appl.494(2021), 124448 (11pp)

2021

Show all 25 references

[6]

P. A. Clarkson and K. Jordaan, Generalised Airy polynomials, J. Phys. A: Math. Theor. 54(2021), 185202 (28pp)

2021
[7]

I. S. Gradshteyn and I. M. Ryzhik, Table of Integrals, Series, and Products, Seventh edition. Edited by Daniel Zwillinger and Victor Moll. Academic Press, Elsevier, 2015

2015
[8]

Ismail, Classical and Quantum Orthogonal Polynomials in One Variable, Encyclopedia of Mathematics and its Applications 98, Cambridge University Press, Cambridge, 2005

M. Ismail, Classical and Quantum Orthogonal Polynomials in One Variable, Encyclopedia of Mathematics and its Applications 98, Cambridge University Press, Cambridge, 2005

2005
[9]

Kang and Q

H. Kang and Q. Xu, Quadrature formulae of many highly oscillatory Frourier-type integrals with algebraic or logarithmic singularities and their error analysis, Appl. Math. Comput. 442(2023), 127758 (15pp)

2023
[10]

D. K. Kurtoˇ glu, A. I. Has¸ celik and G. V. Milovanovi´ c, A method for efficient computation of integrals with oscillatory and singular integrand, Numer. Algorithms85(2020), 1155-1173

2020
[11]

N. N. Lebedev, Special Functions and Their Applications, Revised English Edition Trans- lated and Edited by Richard A. Silverman. Dover Publications, New York, 1972

1972
[13]

G. V. Milovanovi´ c and A. S. Cvetkovi´ c, Orthogonal polynomials related to the oscillatory Chebyshev weight function, Bull. Cl. Sci. Math. Nat. Sci. Math. 30 (2005), 47-60

2005
[16]

G. V. Milovanovi´ c, A. S. Cvetkovi´ c and M. P. Stani´ c, Orthogonal polynomials for modified Gegenbauer weight and corresponding quadratures, Appl. Math. Lett.22(2009), 1189- 1194. 15

2009
[17]

G. V. Milovanovi´ c, M. P. Stani´ c and T. V. Tomovi´ c Mladenovi´ c, Gaussian type quadrature rules related to the oscillatory modification of the generalized Laguerre weight functions, J. Comput. Appl. Math.437(2024), 115476 (8pp)

2024
[18]

Min and Y

C. Min and Y. Chen, Gap probability distribution of the Jacobi unitary ensemble: an elementary treatment, from finitento double scaling, Stud. Appl. Math.140(2017), 202- 220

2017
[19]

Min and Y

C. Min and Y. Chen, Painlev´ e V and the Hankel determinant for a singularly perturbed Jacobi weight, Nucl. Phys. B961(2020), 115221 (25pp)

2020
[20]

Min and Y

C. Min and Y. Chen, Differential, difference and asymptotic relations for Pollaczek-Jacobi type orthogonal polynomials and their Hankel determinants, Stud. Appl. Math.147(2021), 390-416

2021
[21]

Min and P

C. Min and P. Fang, Generalized Airy polynomials, Hankel determinants and asymptotics, Physica D473(2025), 134560 (9pp)

2025
[22]

Min and X

C. Min and X. Wu, Asymptotics of the Hankel determinant and orthogonal polynomials arising from the information theory of MIMO systems, arXiv: 2510.06739

arXiv
[23]

Min and X

C. Min and X. Wu, Orthogonal polynomials for the singularly perturbed Laguerre weight, Hankel determinants and asymptotics, arXiv: 2511.09362
[24]

Mu and S

X. Mu and S. Lyu, Hankel determinants for a deformed Laguerre weight with multiple variables and generalized Painlev´ e V equation, J. Math. Phys.66(2025), 073506 (21pp)

2025
[25]

n", "\[Alpha][n]

M. P. Stani´ c and A. S. Cvetkovi´ c, Orthogonal polynomials with respect to modified Jacobi weight and corresponding quadrature rules of Gaussian type, Numer. Math. Theor. Meth. Appl.4(2011), 478-488. Appendix A Derivation of the initial values{α 0, α1}and{β 1, β2} given by(2...

2011

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [14]

Milovanovi´ c and A

G V. Milovanovi´ c and A. S. Cvetkovi´ c, Orthogonal polynomials and Gaussian quadrature rules related to oscillatory weight functions, J. Comput. Appl. Math. 179 (2005), 263-287

2005

[2] [12]

Lyu and Y

S. Lyu and Y. Lyu, Ladder Operators for Laguerre-type and Jacobi-type Orthogonal Poly- nomials, J. Phys. A: Math. Theor.59(2026), 115201 (30pp)

2026

[3] [15]

G. V. Milovanovi´ c, A. S. Cvetkovi´ c and Z. M. Marjanovic, Orthogonal polynomials for the oscillatory-Gegenbauer weight, Publ. Inst. Math. (Belgr.) (N.S.)84(2008), 49-60

2008

[4] [1]

Basor, Y

E. Basor, Y. Chen and N. Haq, Asymptotics of determinants of Hankel matrices via non- linear difference equations, J. Approx. Theory198(2015), 63-110

2015

[5] [2]

Chen and M

Y. Chen and M. Ismail, Jacobi polynomials from compatibility conditions, Proc. Am. Math. Soc.133(2005), 465-472

2005

[6] [3]

Chen and A

Y. Chen and A. Its, Painlev´ e III and a singular linear statistics in Hermitian random matrix ensembles, I, J. Approx. Theory162(2010), 270-297

2010

[7] [4]

Chen and M

Y. Chen and M. McKay, Coulomb fluid, Painlev´ e transcendents and the information theory of MIMO systems, IEEE Trans. Inf. Theory58(2012), 4594-4634

2012

[8] [5]

R. Chen, D. Yu and J. Chen, Computation of oscillatory integrals with an exponential kernel and Jacobi-type singularities, J. Math. Anal. Appl.494(2021), 124448 (11pp)

2021