Pith. sign in

REVIEW 4 major objections 3 minor 11 references

On completely monotonic functions

T0 review · 4 major / 3 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 characterizes exactly which polynomial multiples of a base difference stay completely monotonic: the multiplier must be $x+c_0$ with $c_0$ above an explicit bound.

desk verdict The paper's central theorem is not just unproved but false, and the counterexample is immediate. read the letter →

arxiv 2505.23767 v1 pith:EDBXE4Z2 submitted 2025-05-29 math.CA

Mostafa Najafi , Ali Morassaei This is my paper
classification math.CA MSC 26A4826D0714F1030E20
keywords completelymonotonicfunctionsBernsteinBernstein'stheoremStieltjestransformLaplacerepresentationpolynomialmultiplierscompletemonotonicitythreshold
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

Completely monotonic functions are smooth functions on the positive half-line whose derivatives alternate in sign; by Bernstein's theorem they are exactly the Laplace transforms of nonnegative measures. The paper studies whether multiplying a particular base difference, $e^{ab}-(1+a/x)^{bx+d}$, by a real polynomial preserves this property. Its main theorem gives a sharp answer: for $P_n(x)=x^n+\sum_{i=0}^{n-1}c_i x^i$, the product is completely monotonic if and only if $n=1$ and $c_0$ is at least the explicit threshold $c=\frac{a}{12}\left(6+ab\left(3+\frac{2}{ab-2d}\right)-6d\right)$. The interest is that this converts a function-theoretic property into a single coefficient inequality and produces a large family of explicit completely monotonic functions parameterized by $a,b,d$ and $c_0$.

What carries the argument

The carrying object is the Stieltjes-to-Laplace representation chain. Lemma 2.33 expresses the base difference, after the prefactor $x+1$, as a constant plus the Stieltjes transform of the signed density $$g(s)=\frac{ad}{\pi}$s^{{bs-d}}$(1-s)^{-bs+d+1}\sin(\pi s)\quad(0\le s\le1).$$ Substituting the exponential identity for $1/(x+s)$ turns this into $\int_0^\infty e^{-xt}h(t)\,dt$ with $h(t)=\int_0^1 e^{-st}g(s)\,ds$. The factor $(x+c)/(x+1)$ that relates $f_c$ to this base representation is then absorbed using the convolution theorem for Laplace transforms, producing the new kernel $u(t)$ in (2.20). The sufficiency of the coefficient condition is exactly the claim that this $u(t)$ is nonnegative for $t>0$.

What would settle it

Evaluate $u(t)$ defined by (2.20) numerically for a parameter choice with $0<c<1$, for instance $a=1$, $b=1$, $d=1/4$ (where $c=23/24$), across a range of $t$. If any parameter triple and $t$ give $u(t)<0$, the claimed sufficiency direction is false; the same parameters also make the displayed inequality (2.22) fail for every $s\in(c,1)$, so this is the step to test.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.35: for real $a,b,d$ (with the expression defined, so $ab\neq2d$) and a real polynomial $P_n$ of degree $n\ge1$, the function $x\mapsto P_n(x)\left(e^{ab}-(1+a/x)^{bx+d}\right)$ is completely monotonic on $(0,\infty)$ exactly when $n=1$ and $c_0\ge \frac{a}{12}\left(6+ab\left(3+\frac{2}{ab-2d}\right)-6d\right)$. The proof of the 'if' direction represents the base factor as a Stieltjes integral, $$(x+1)\left($e^{{ab}}$-(1+\tfrac{a}{x})^{bx+d}\right)=\tfrac12 a(ab-2d)$e^{{ab}}$+\frac{ad}{\pi}\$int_0^{1}$ \frac{$s^{{bs-d}}$(1-s)^{-bs+d+1}\sin(\pi s)}{x+s}\,ds,$$ then uses the identity $1/(x+s)=\int_0^\infty e^{-xt}e^{-st}dt$ to pass to a Laplace representation whose kernel $u(t)$ is shown to be nonnegative whenever $c_0$ lies above the threshold. The 'only if' direction runs through the asymptotic ratio $xP_n'(x)/P_n(x)$, which forces $n=1$, and then the same limit pins down the required lower bound on $c_0$.

Load-bearing premise

The sufficiency proof depends on the comparison $e^{-st}(c-s)\ge e^{-ct}c$ for $t>0$ and $s\in(c,1)$ from Eq. (2.22); this is what forces the kernel $u(t)$ to be nonnegative, and if the comparison fails the complete-monotonicity conclusion does not follow.

Editorial extensions

If this is right

  • For fixed $a,b,d$ with $ab\neq2d$, the admissible polynomial multipliers form a half-line: $P_1(x)=x+c_0$ with $c_0$ at least the displayed threshold.
  • Every function in this admissible family is completely monotonic, so each is the Laplace transform of a nonnegative measure; the construction gives the kernel $u(t)$ explicitly.
  • A polynomial factor of degree $n\ge2$ never works, no matter how its coefficients are chosen.
  • The threshold is an explicit rational function of the parameters, so the boundary of the admissible region can be evaluated directly for any allowed triple $(a,b,d)$.

Reading between the lines

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

  • The same Stieltjes-density machinery would apply to other differences of the form $e^{A}-(1+a/x)^{Bx+C}$; only the moment integrals $\int_0^1 s^k g(s)\,ds$ would need to be recomputed, so the approach is a template rather than a one-off computation.
  • Because the admissible multipliers are exactly the linear polynomials $x+c_0$, the theorem gives a natural way to separate polynomial corrections that preserve complete monotonicity from those that destroy it; this could be useful in approximation problems where sign-alternating derivatives are a design constraint.
  • The parameter $d$ is essentially free in the threshold, away from the pole $ab=2d$, so holding $a,b$ fixed and varying $d$ lets one tune how large $c_0$ must be; the paper does not map out this transition, but the formula makes it checkable.
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 / 3 minor

Summary. The manuscript surveys closure properties of completely monotonic functions and then focuses on characterizing real polynomials P_n for which x ↦ P_n(x)(e^{ab} - (1 + a/x)^{bx+d}) is completely monotonic. The central result, Theorem 2.35, claims that this holds if and only if n = 1 and the constant coefficient c0 is at least c = (a/12)(6 + ab(3 + 2/(ab-2d)) - 6d). The sufficiency proof combines a Stieltjes-type representation (Lemma 2.33), an integral evaluation attributed to Alzer-Berg (Lemma 2.34), and a Laplace-transform positivity argument. The paper also states numerous auxiliary lemmas and corollaries about products, powers, and compositions of completely monotonic and Bernstein functions.

Significance. If correct, Theorem 2.35 would give a sharp degree-and-coefficient criterion for a natural parametric family, extending earlier work of Alzer and Berg, and would merit publication. However, the central theorem is false as stated: a simple counterexample with a=b=d=1 and c0=1 satisfies the claimed inequality but the function is negative for large x. The proof also contains an invalid inequality at the key positivity step, and Lemma 2.34 has an internal sign inconsistency. These are not local or cosmetic issues; they invalidate the advertised characterization. The auxiliary material includes further false claims, such as Corollary 2.11, so the paper in its current form cannot be accepted.

major comments (4)
  1. [Theorem 2.35] The statement is false as written. Take a=b=d=1, n=1, and c0=1. The threshold is c=(1/12)(6+3-2-6)=1/12, so the condition c0 ≥ c holds. But F(x)=(x+1)(e-(1+1/x)^{x+1}) satisfies (x+1)log(1+1/x)=1+1/(2x)-1/(6x^2)+O(x^{-3}), hence (1+1/x)^{x+1}=e(1+1/(2x)-1/(24x^2)+O(x^{-3})) and F(x)=-e/2 - 11e/(24x)+O(x^{-2}). Thus F(x)<0 for all sufficiently large x, contradicting the requirement (-1)^0 F(x) ≥ 0 in the definition of complete monotonicity. This also shows that the constant term in Lemma 2.33, (1/2)a(ab-2d)e^{ab} = -e/2, is negative, so the claimed Stieltjes representation cannot hold with a nonnegative measure.
  2. [Eq. (2.22)] The inequality at Eq. (2.22) is invalid. For s ∈ (c,1), the factor (c-s) is negative, and the step replacing e^{-st}(c-s) by e^{-ct} c does not give a lower bound; the correct lower bound in that range would be e^{-ct}(c-s), which is negative. Consequently the claimed bound w'(t)e^{-t} ≥ e^{-ct} ∫_0^1 (c-s)g(s) ds is not established. Since this inequality is the only argument showing that u(t) ≥ 0, the conclusion that f_c is completely monotonic from the representation (2.20) does not follow.
  3. [Lemma 2.34] Lemma 2.34 is internally inconsistent. For a=b=d=1, the function g(s) in Eq. (2.6) is strictly positive on (0,1), since sin(πs)>0 and the powers are positive. Lemma 2.34 nevertheless yields ∫_0^1 g(s) ds = -11e/24 < 0. The proof of Theorem 2.35 uses this value to compute ∫_0^1 (c-s)g(s) ds and to conclude that w(0)=0, so the numerical input to the sufficiency argument is wrong.
  4. [Corollary 2.11] Corollary 2.11 is false. If f(x)=x^2, a=e, and t=1, the corollary asserts that e^{-x^2} is completely monotonic on (0,∞). But (e^{-x^2})''(0)=-2<0, so the required condition (-1)^2 f''(0) ≥ 0 fails. The proof invokes Theorem 2.1(iii), which requires f to be a Bernstein function; x^2 is not a Bernstein function.
minor comments (3)
  1. [Corollary 2.10] The proof of Corollary 2.10 does not follow from Theorem 2.9, because Theorem 2.9 requires the inner function to map into a fixed interval where the outer function is absolutely monotonic, which is not guaranteed for an arbitrary completely monotonic function. A correct proof would need Faa di Bruno's formula or a different composition theorem.
  2. [Example 2.20(iv)] Example 2.20(iv) asserts that (1+x)^{1/x} is completely monotonic on (0,∞) without proof. This is not an immediate consequence of the cited results and requires justification or removal.
  3. [Throughout] There are numerous typographical and presentational issues, including misspellings such as 'absouletly', 'obviouse', and inconsistent references such as 'Theorem 2.10' when Corollary 2.10 is intended. These should be corrected in any revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation uses quoted external results (Alzer–Berg, Lemma 2.34) and algebraic identities, rather than assuming or repackaging its own conclusion.

full rationale

I examined the derivation chain leading to Theorem 2.35. The necessity argument uses the limit of the auxiliary function g_n in (2.9) and the inequality (2.10); this is an asymptotic analysis, not an assumption of the conclusion. The sufficiency argument reduces the case of general c0 to the case c0 = 1 via the exact identity fc0 = fc + (c0 - c)/(x + c) fc and the product/composition closure properties in Remark 2.4, then represents fc as a Laplace transform with kernel u(t) defined in (2.20). Nothing in this chain fits a parameter to the target function or defines the conclusion into the premises. The paper leans heavily on Lemma 2.34, but that lemma is attributed to Alzer and Berg [2], an external source, and it is used as an integral evaluation, not as a disguised restatement of Theorem 2.35. There are no self-citations by the present authors. The main mathematical concern is the invalid estimate in equation (2.22), where for s in (c,1) the inequality e^{-st}(c-s) >= e^{-ct} c is not true, and the claimed positivity of w'(t)e^{-t} therefore does not follow. That is a correctness defect, not a circularity: it is a broken intermediate estimate, not a reduction of the theorem to its own assertion. Likewise, the apparent inconsistency for a=b=d=1 between Lemma 2.33, Lemma 2.34, and Theorem 2.35 is an internal mathematical error rather than a circular dependence. Because no load-bearing step is equivalent by construction to the paper's inputs, the circularity score is 0.

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

The paper introduces no free parameters or new entities; it relies on standard results in complete monotonicity and on the Alzer-Berg integral identities quoted as Lemma 2.34.

assumptions (4)
  • standard math Bernstein's theorem: f is completely monotonic iff f(x) = ∫ e^{-xt} dμ(t) for a nonnegative measure μ
    Used throughout as the fundamental characterization; cited from Widder [10].
  • standard math Stieltjes transform representation criterion used in Lemma 2.33
    Invoked to represent the base function; standard result from Akhiezer [1].
  • domain assumption Integral identities of Lemma 2.34 from Alzer-Berg [2]
    The central proof uses these quoted identities; the paper does not derive them.
  • standard math Product and composition rules for completely monotonic functions (Lemma 2.3, Theorem 2.1)
    Used systematically in the 'trivial observations' and in Theorem 2.35.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On completely monotonic functions." pith.science (2026). https://pith.science/paper/EDBXE4Z2

@misc{pith2026250523767,
  author       = {Pith},
  title        = {Pith review of: On completely monotonic functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EDBXE4Z2}},
  note         = {Machine review of arXiv:2505.23767}
}
abstract

Let $ f:(0,\infty)\rightarrow \Bbb{R} $ be a completely monotonic function. In this paper, we present some properties of this functions and several new classes of completely monotonic functions. We also give some special functions such that its have completely monotonic condition.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

  1. [1]

    Akhiezer, The Classical Moment Problem and Some Related Questions in A nalysis

    N.I. Akhiezer, The Classical Moment Problem and Some Related Questions in A nalysis. - English translation, Oliver and Boyd, Edinburgh, 1965

    work page 1965

  2. [2]

    Alzer, C

    H. Alzer, C. Berg, Some classes of completely monotonic function s, Ann. Acad. Sci. Fenn. Math. 27 (2002), 445–460

    work page 2002

  3. [3]

    Alzer, C

    H. Alzer, C. Berg, Some classes of completely monotonic function s. II. Ramanujan J. 11 (2006), no. 2, 225-248

    work page 2006

  4. [4]

    Bochner, Harmonic analysis and the theory of probability , Univ

    S. Bochner, Harmonic analysis and the theory of probability , Univ. of California Press, Berkeley-Los Angeles, 1960

    work page 1960

  5. [5]

    Doob, Classical Potential Theory and its Probabilistic Counterp art

    J.L. Doob, Classical Potential Theory and its Probabilistic Counterp art. - Springer- Verlag, New York, 1984

    work page 1984

  6. [6]

    Gradshtein and I.M

    I.S. Gradshtein and I.M. Ryzhik, Tables of integrals, Sums, Series and products . 7th Edition, Academic Press, Inc., 1994. ON COMPLETELY MONOTONIC FUNCTIONS 15

    work page 1994

  7. [7]

    Kimberling, A probabilistic interpretation of complete monoton icity, Aequat

    C.H. Kimberling, A probabilistic interpretation of complete monoton icity, Aequat. Math. 10 (1974), 152–164

    work page 1974

  8. [8]

    Rosser, The complete monotonicity of certain functions der ived from completely mono- tonic functions

    J.B. Rosser, The complete monotonicity of certain functions der ived from completely mono- tonic functions. Duke Math. J. 15 (1948), 313–331

    work page 1948

Show all 11 references
  1. [9]

    R. L. Schilling, R. Song, Z. Vondraˇ cek, Bernstein functions: Theory and applications . 2012

    2012

  2. [10]

    Widder, The Laplace transform

    D.V. Widder, The Laplace transform . Princeton Univ. Press, Princeton, NJ , 1941

    1941

  3. [11]

    Zhang, C

    S. Zhang, C. Chen, F. Qi, On a completely monotonic function. Math. Practice Theory, 36 (2006) , 236-238. 1,2Department of Mathematics, F aculty of Sciences, Universit y of Zanjan, Uni- versity Blvd., Zanjan 45371-38791, Iran. Email address : mostafa.najafi@znu.ac.ir Email add...

    2006

Pith tools

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