Generalized product formulas for Whittaker's functions and a novel class of index transforms

Semyon Yakubovich

REVIEW 5 major objections 5 minor 11 references

Generalized product formulas for Whittaker's functions and a novel class of index transforms

T0 review · 5 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper proves generalized product formulas for Whittaker functions with different indices and derives a new class of index transforms with an explicit inversion formula.

desk verdict Real extension of the author's index-transform theory, with an explicit inversion theorem, but the proof's central L2 condition (4.13) is asserted rather than proved. read the letter →

arxiv 2506.05013 v2 pith:7VOIU4SK submitted 2025-06-05 math.CA

Semyon Yakubovich This is my paper

classification math.CA MSC 45A0544A1533C1533C10

keywords WhittakerfunctionindextransformOlevskiiKontorovich-LebedevMellin-BarnesintegralproductformulainversionmodifiedBessel

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 build an index-transform theory for products of Whittaker functions whose two Whittaker indices differ. It proves product formulas that represent such products as Laplace-type integrals with Gauss hypergeometric and Appell-function kernels, then introduces the index transform $F^{\nu}_{\alpha,\beta}$ of (1.26). The central result is an absolutely convergent inversion formula for $F^{\nu}_{\alpha,\beta}f$ with an explicitly written double-series kernel, given in Theorem 4. The new reciprocal pairs specialize to known Kontorovich-Lebedev, Olevskii, and Lebedev transforms, so the work extends a classical chain of integral-transform pairs.

What carries the argument

The load-bearing mechanism is the product formula (1.25), which rewrites the kernel of $F^{\nu}_{\alpha,\beta}$ as a Laplace integral whose integrand is the Gauss hypergeometric kernel of a generalized Olevskii transform of complex index $2\nu+i\tau$. This decomposes the index transform as a Laplace transform followed by a generalized Olevskii operator (Theorem 2). The inversion chain then runs through Mellin transform identities, the replacement of a vertical Mellin-Barnes contour by a left-hand loop, and an injectivity argument for the Laplace transform, terminating in the double hypergeometric series $\Psi^{\alpha,\beta}_{\nu,i\tau}(t)$ of (4.23).

What would settle it

With $\nu=0.05$, $\alpha=-0.15$, $\beta=0.05$ and $f(\tau)=e^{-\tau^2}$, compute $F^{\nu}_{\alpha,\beta}f$ by (1.26) to high precision, then evaluate the right side of (4.22) using the double series (4.23) truncated at increasing orders; if the reconstruction error does not tend to zero as the truncation and quadrature resolution grow, the inversion formula as stated is false.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the product $W_{\alpha,-2\nu-i\tau}(x)W_{\beta,\nu+i\tau}(x)$ can be represented as a Laplace transform of a generalized Olevskii kernel, making the index transform (1.26) a composition of an ordinary Laplace transform and a generalized Olevskii transform. The author inverts both factors, using $L^2$ Parseval equalities for Mellin transforms, a contour replacement from a vertical Mellin-Barnes line to a left-hand loop, and Laplace inversion. The outcome is Theorem 4: for $f\in L^2(\mathbb{R})$, $\nu\neq 0$, $\nu+\alpha+\beta<0$, $\beta-\alpha>\max\{\nu,-3\nu\}$, the inversion formula $f(\tau)=\frac{1}{\pi}\int_0^\infty e^{-t}t^{2(\beta-\alpha)-1}\Psi^{\alpha,\beta}_{\nu,i\tau}(t)(F^{\nu}_{\alpha,\beta}f)(t)\,dt$ holds with $\Psi$ given explicitly by the absolutely convergent double series (4.23).

Load-bearing premise

The inversion formula rests on several interchanges of limits and integrals plus an asserted square-integrability condition on a certain Mellin-Barnes integral; if any of those steps fails for allowed parameters, the explicit kernel $\Psi$ would not be validated.

Editorial extensions

If this is right

Any $f\in L^2(\mathbb{R})$ in the stated parameter range is recoverable exactly from its index-transform data through (4.22), making the pair reciprocal.
Particular parameter choices reproduce the Kontorovich-Lebedev pair, the classical Olevskii transform, the Lebedev transform with the square of the Macdonald function, and earlier products-of-Whittaker cases.
The product formulas (1.8)-(1.25) provide new integral representations for products of Whittaker functions with different indices, giving kernels for further integral operators.
The $\nu=0$ inversion formula (4.24) covers transforms previously studied only in special cases, including the square-of-Whittaker index transform.

Reading between the lines

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

The parameter restrictions in Theorem 4 are tight (for $\nu>0$ they imply $\alpha<-\nu$), so a natural next step is analytic continuation to the boundary cases $\nu=0$ and $\nu+\alpha+\beta=0$ that the paper handles by separate formulas.
Because $\Psi$ is an absolutely convergent double series, the inversion formula is numerically evaluable; testing it on rapidly decaying functions would map its practical range.
The Laplace-Olevskii composition suggests that a convolution theory for this new index transform could be built along the lines of the equal-index case, but the paper does not develop that step.

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

Yakubovich has a genuine extension here: new product formulas for products of Whittaker functions with different indices, and a new two-parameter family of index transforms F^ν_{α,β} with an explicit inversion formula (Theorem 4). The paper is worth refereeing, but the proof has a load-bearing gap at condition (4.13).

What is new: the product representations (1.14)-(1.23) and the two-index inversion with kernel Ψ in (4.23) are not in the earlier papers [8], [9], [11] or Wimp. The reductions to Olevskii, Kontorovich-Lebedev, Hankel and Mellin transforms are skillfully arranged, parameter conditions are stated explicitly, and several convergence estimates are genuinely checked. The L2 norm bound for the generalized Olevskii transform (2.9) is a useful byproduct. The paper does what good classical analysis should do: it reduces a new object to known machinery without fitted parameters or circularity.

The soft spots are real but concentrated. The step from the double Mellin-Barnes integral (4.11) to the single integral (4.14) requires the inner w-integral to be in L2 on the vertical line (4.13). The author asserts this after shifting the w-contour to a left-hand loop, with only one sentence about Stirling asymptotics. The shift is s-dependent because the poles of Γ(w+(ν+β−α+s)/2−1) and its companion move with s, so the estimate has to be uniform in s; no such estimate is shown. If this fails, the explicit kernel Ψ in (4.23) is not connected to the transform (1.26). That is a genuine gap, not a cosmetic one. Relatedly, the absolute convergence of the double series defining Ψ is delegated to conditions (8.3) of [10] without checking those conditions against the theorem's hypotheses. The author also explicitly leaves the M-function product representation (4.28) to the interested reader; that is a special case, so it is not load-bearing for Theorem 4, but it is a sign that some contour and series reductions are unfinished.

The citation pattern is self-centric but understandable: the author is building on his own prior theory, and I do not see an outside source being ignored. The conditional verdict from the reader is fair; this is not a desk-reject candidate.

Who this is for: specialists in integral transforms and special functions. A serious referee should be assigned, with the specific instruction to verify (4.13) and the s-dependent contour shift, and to confirm that the [10] conditions cover the double series. I expect the theorem is true and the gaps fixable, but as written the central justification is incomplete.

Referee Report

5 major / 5 minor

Summary. The paper establishes generalized product formulas for products of Whittaker functions with different indices and uses them to study the index transform (1.26). The main results are: a composition representation of the transform in terms of a generalized Olevskii transform and the Laplace transform (Theorem 2), an L2-boundedness and inversion result for the generalized Olevskii transform (Theorem 3), and an explicit inversion formula for the original transform with a double-series kernel Psi (Theorem 4, equations (4.22)-(4.23)). Particular cases recover index transforms with products of Macdonald functions and with products of Whittaker M-functions.

Significance. If Theorem 4 is correct, the paper gives a genuinely new reciprocal pair of integral transforms whose kernel is an explicit infinite series in products of Whittaker functions, extending Wimp's and Kontorovich-Lebedev type theories. The proof strategy is attractive: it reduces the inversion to the Olevskii, Kontorovich-Lebedev, and Mellin transforms and supplies norm estimates for the generalized Olevskii operator. The paper is careful about many convergence conditions and dominated-convergence interchanges. However, two load-bearing analytic steps are not actually proved: the L2 condition (4.13) needed for the Mellin-Parseval reduction, and the absolute convergence of the double series defining the explicit kernel Psi in (4.23). Because both steps are central to the inversion theorem, the main claim is not yet fully established, although it appears plausible and likely repairable.

major comments (5)

[Section 4, condition (4.13)] The passage from the double Mellin-Barnes representation (4.11) to the single-integral formula (4.14) uses the Mellin-Parseval equality and therefore requires that the inner w-integral in (4.13) belong to L2 on the vertical line Re(s)=1-nu-alpha-beta-gamma. The paper asserts this after replacing the vertical w-contour by the left-hand loop L_-infinity, saying that 'the Stirling asymptotic formula for the gamma function' gives the condition. This is not shown. The deformation is s-dependent because the poles of Gamma(w+(nu+beta-alpha+s)/2-1) and Gamma(w+(nu+beta-alpha+s-1)/2) move with s, so one must prove that the loop integral is controlled uniformly in s, including possible residue contributions that may grow in Im(s). Without a detailed estimate of the resulting function of s on the line, the explicit kernel Psi in (4.23) has not been rigorously connected to the transform (1.26). This is the main load-bearing gap.
[Section 4, equations (4.21) and (4.23)] The absolute convergence of the double series defining Psi is delegated to 'conditions (8.3)' of reference [10], but those conditions are not checked against the hypotheses nu+alpha+beta<0 and beta-alpha>max{nu,-3nu}. The kernel Psi contains terms of the form t^{6nu+2i tau} and t^{-2(nu+i tau)}, so convergence near t=0 and t=infinity is not automatic and depends on the sign of nu. Moreover, the gamma functions in the denominators can have poles for particular parameter values, and no generic-position assumptions are stated. Since Theorem 4 asserts absolute convergence of the inversion integral (4.22), this is a necessary part of the proof and cannot be left as a reference to unverified conditions.
[Section 4, Theorem 4 statement] The statement of Theorem 4 says 'under conditions of Theorems 2, 3' without listing those conditions, and the proof additionally uses a nonempty interval for gamma, namely max{1/2, 1/4-nu+(alpha+beta)/2, -nu-alpha-beta, nu+beta-3alpha/2} < gamma < min{nu+beta-alpha-1/2, 3/4-alpha+beta/2+nu/2}. It is not shown that this interval is nonempty under the hypotheses of the theorem. In addition, the inversion formula (3.13) for the generalized Olevskii transform requires the extra intersection condition (3.12), including an L1 membership; the proof derives or implies this, but the theorem statement should either include the full hypotheses or state explicitly which conditions are assumed and which are proved.
[Section 4, equation (4.16)] The interchange of integrations leading to (4.16) is justified by the asymptotic bounds (3.7)-(3.9) together with the bound |Phi| = O(x^{nu+alpha+beta+gamma-1}). The displayed estimate contains an integral over x in (0,1) with denominator (1-x)^{nu+alpha+beta+1}; since nu+alpha+beta<0 this is integrable, but the displayed chain of inequalities is written in a compressed form and the reader is not told which gamma-conditions make the outer t-integral converge. This is probably fixable, but it should be written out because it is part of the absolute-convergence claim of Theorem 4.
[Section 4, equation (4.28)] The reduction of the kernel to a product of Whittaker M-functions for alpha=-1/4 is explicitly said to hold 'under some justifications which we leave for the interested reader.' This is not load-bearing for Theorem 4, but it is presented as a derived formula; it should either be proved or clearly labeled as a conjecture or as an auxiliary computation whose proof is omitted.

minor comments (5)

[Section 2, Lemma 2] The hypothesis of Lemma 2 is f in L2((-1,1); d tau / tau^2) intersect L2(R\setminus(-1,1); d tau); this is a nonstandard space and the relation to the usual condition f in L2(R) is not discussed. The statement would be easier to use if the exact domain were described in the theorem that relies on it.
[Section 1, equation (1.6)] The citation '[3, formula (3.19)' is missing a closing parenthesis; it should read '[3, formula (3.19)]'.
[Throughout] There are several OCR-type typographical artifacts, for example 'Whittaker ~Os function' in reference [11] and the repeated '))' in the denominator of (4.7). A careful proofreading pass is needed.
[Section 4, equation (4.12)] The notation I_-^{nu+alpha+beta} for the Riemann-Liouville fractional integral is used without an explicit definition, although the preceding displayed formula makes the meaning clear. Adding a consistent definition would improve readability.
[Section 4, equation (4.20)] The inversion of the Laplace transform via Entry 3.35.2.5 in [2, Vol. V] is invoked without stating the parameter conditions under which this entry applies. The relevant inequalities should be made explicit for the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the inversion theorem is derived from standard product and Olevskii-transform identities, not from its own conclusion.

full rationale

The paper contains a genuine derivation chain: the product formula (1.25) is obtained from known integral representations, the transform (1.26) is expanded as a composition of Laplace and generalized Olevskii transforms in (3.5), and the inversion formula (4.22) is obtained by inverting that composition via Mellin-Barnes contours and hypergeometric series. No parameter is fitted to the target result, no prediction is defined by the conclusion, and no uniqueness theorem is imported from the author's prior work to force the chosen kernel. The kernel (4.23) is not assumed but is produced by explicit contour evaluation and series manipulation. The main weakness is a proof gap: condition (4.13) is asserted after a one-sentence Stirling-asymptotic justification, and the special-case reduction (4.28) is explicitly left for the reader. These are gaps in rigor, not circularity, because the missing estimates do not assume the inversion formula being proved. The self-citations, including [3], [7], [8], [9], [10], and [11], refer to independently published books and papers used as standard lemmas or as context for particular cases; none of them is the sole justification of the central claim, and the cited results do not reduce to the present theorem by construction. The Lebedev-transform particular case is explicitly identified as a known result, not presented as novel. The paper's central content is therefore self-contained in the relevant sense, and no circular step can be exhibited with a quoted reduction.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The derivation relies entirely on established special-function identities and analytic techniques. No free parameters are fitted; α, β, ν and τ are transform parameters, not hand-tuned constants. The main fragility is a set of analytic-continuation and interchange assertions, including an explicit 'left to the interested reader' step in (4.28).

assumptions (6)

standard math Whittaker function has the Laplace integral representation (1.18).
Invoked in Section 1 as the departure point for the convolution representation of the product; standard and independently checkable.
standard math Macdonald function product formula (1.12).
Used to pass from Macdonald products to a single kernel; taken from Prudnikov et al. handbook.
standard math Olevskii kernel integral representation (2.2) with Bessel and Macdonald functions.
Foundation of Section 2; from the author's earlier paper [9] and a standard handbook entry.
standard math Mellin-Barnes representation (4.6) for the Gauss hypergeometric function with factor (x(x+2))^{-(ν+α+β)}.
Used to derive the double Mellin-Barnes form (4.11) and the explicit kernel (4.23); from handbook Entry 8.4.49.24.
domain assumption All interchanges of integration in Sections 2 through 4 are valid under the stated decay and L2 conditions.
The assertions rely on dominated convergence, Fubini and Parseval; the paper provides only outline estimates in several places.
ad hoc to paper Formula (4.28) expressing the inverse kernel via products of Whittaker M-functions under α=−1/4.
This is an explicit unproved assertion in the text, stated 'under some justifications which we leave for the interested reader'.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Generalized product formulas for Whittaker's functions and a novel class of index transforms." pith.science (2026). https://pith.science/paper/7VOIU4SK

@misc{pith2026250605013,
  author       = {Pith},
  title        = {Pith review of: Generalized product formulas for Whittaker's functions and a novel class of index transforms},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7VOIU4SK}},
  note         = {Machine review of arXiv:2506.05013}
}

read the original abstract

Generalized product formulas and index transforms, involving products of Whittaker's functions of different indices are established and investigated. The corresponding inversion formulas are found. Particular cases cover index transforms with products of the modified Bessel and Whittaker's functions. For our goals the Kontorovich-Lebedev and Olevskii transforms of a complex index with nonzero real part are involved.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

[8]

Yakubovich, Index transforms associated with products of Whittaker’s functions

S. Yakubovich, Index transforms associated with products of Whittaker’s functions. J. Comput. Appl. Math. Vol. 148 (2002), no. 2, 419-427. 44 Semyon YAKUBOVICH

work page 2002
[9]

Yakubovich, On the Plancherel theorem for the Olevskii transform

S. Yakubovich, On the Plancherel theorem for the Olevskii transform. Acta Math Vietnam. Vol. 31 (2006), 249-260

work page 2006
[11]

A new index transform with the square of Whittaker's function

S. Yakubovich, A new index transform with the square of Whittaker ˜Os function. ArXiv:2503.23770. S.Yakubovich Department of Mathematics, Faculty of Sciences, University of Porto, Campo Alegre st., 687 4169-007 Porto Portugal E-Mail: syakubov@fc.up.pt

work page Pith review arXiv
[10]

Yakubovich (1992)

Nguyen Thanh Hai and S. Yakubovich (1992). The double Mellin-Barnes type integrals and their applications to convolution theory.World Scientific Intern. Pub!. Series on Soviet and East European Mathematics . Vol. 6. Singapore, New Jersey, London and Hong Kong, 1992

work page 1992
[1]

Yu. A. Brychkov, O.I. Marichev and N.V. Savischenko, Handbook of Mellin trans- forms. Advances in Applied Mathematics. CRC Press, Boca Raton, FL, 2019

work page 2019
[2]

Prudnikov, Yu.A

A.P. Prudnikov, Yu.A. Brychkov and O.I. Marichev, Integrals and Series . Vol. I: Elementary Functions, Vol. II: Special Functions, Gordon and Breach, New York and London, 1986, Vol. III : More special functions, Gordon and Breach, New York and London, 1990, Vol. V : Inverse Laplace transforms , Gordon and Breach, New York and London, 1992

work page 1986
[3]

Sousa, M

R. Sousa, M. Guerra and S. Yakubovich, Convolution-like structures, differential operators and diffusion processes . Lecture Notes in Mathematics, 2315. Springer, Cham, 2022

work page 2022
[4]

Titchmarsh, Introduction to the theory of Fourier integrals , Third edition

E.C. Titchmarsh, Introduction to the theory of Fourier integrals , Third edition. Chelsea Publishing Co., New York, 1986

work page 1986

Show all 11 references

[5]

Wimp, A class of integral transforms, Proc

J. Wimp, A class of integral transforms, Proc. Edinb. Math. Soc. 14 (1964), N 2, 33-40

1964
[6]

Yakubovich and Yu

S. Yakubovich and Yu. Luchko, The Hypergeometric Approach to Integral Trans- forms and Convolutions, Kluwer Academic Publishers, Mathematics and Applica- tions. Vol.287, Dordrecht, 1994

1994
[7]

Yakubovich, Index Transforms, World Scientific Publishing Company, Singapore, New Jersey, London and Hong Kong, 1996

S. Yakubovich, Index Transforms, World Scientific Publishing Company, Singapore, New Jersey, London and Hong Kong, 1996

1996

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [8]

Yakubovich, Index transforms associated with products of Whittaker’s functions

S. Yakubovich, Index transforms associated with products of Whittaker’s functions. J. Comput. Appl. Math. Vol. 148 (2002), no. 2, 419-427. 44 Semyon YAKUBOVICH

work page 2002

[2] [9]

Yakubovich, On the Plancherel theorem for the Olevskii transform

S. Yakubovich, On the Plancherel theorem for the Olevskii transform. Acta Math Vietnam. Vol. 31 (2006), 249-260

work page 2006

[3] [11]

A new index transform with the square of Whittaker's function

S. Yakubovich, A new index transform with the square of Whittaker ˜Os function. ArXiv:2503.23770. S.Yakubovich Department of Mathematics, Faculty of Sciences, University of Porto, Campo Alegre st., 687 4169-007 Porto Portugal E-Mail: syakubov@fc.up.pt

work page Pith review arXiv

[4] [10]

Yakubovich (1992)

Nguyen Thanh Hai and S. Yakubovich (1992). The double Mellin-Barnes type integrals and their applications to convolution theory.World Scientific Intern. Pub!. Series on Soviet and East European Mathematics . Vol. 6. Singapore, New Jersey, London and Hong Kong, 1992

work page 1992

[5] [1]

Yu. A. Brychkov, O.I. Marichev and N.V. Savischenko, Handbook of Mellin trans- forms. Advances in Applied Mathematics. CRC Press, Boca Raton, FL, 2019

work page 2019

[6] [2]

Prudnikov, Yu.A

A.P. Prudnikov, Yu.A. Brychkov and O.I. Marichev, Integrals and Series . Vol. I: Elementary Functions, Vol. II: Special Functions, Gordon and Breach, New York and London, 1986, Vol. III : More special functions, Gordon and Breach, New York and London, 1990, Vol. V : Inverse Laplace transforms , Gordon and Breach, New York and London, 1992

work page 1986

[7] [3]

Sousa, M

R. Sousa, M. Guerra and S. Yakubovich, Convolution-like structures, differential operators and diffusion processes . Lecture Notes in Mathematics, 2315. Springer, Cham, 2022

work page 2022

[8] [4]

Titchmarsh, Introduction to the theory of Fourier integrals , Third edition

E.C. Titchmarsh, Introduction to the theory of Fourier integrals , Third edition. Chelsea Publishing Co., New York, 1986

work page 1986