betaFeedback

WorksMathematics

Low moments of sums of extended Rademacher random multiplicative functions

Abstract

Let Xe ⁣:N→{±1}X^e \colon \mathbb{N} \to \{\pm 1\} be the random, completely multiplicative function whose values at the primes (Xe(p))p prime(X^e(p))_{p \text{ prime}} are independent and uniformly distributed on {±1}\{ \pm 1 \}. Let X=Xeμ2X=X^eμ^2. We determine the order of magnitude of E[∣1x∑n≤xXe(n)∣2q] and E[∣∑n≤xX(n)n∣2q] \mathbb{E} \Bigg[ \bigg| \frac{1}{\sqrt{x}}\sum_{n \le x} X^e(n) \bigg|^{2 q} \Bigg]\quad \text{ and }\quad \mathbb{E} \Bigg[ \bigg|\sum_{n \le x} \frac{X(n)}{\sqrt{n}} \bigg|^{2 q} \Bigg] for q∈[0,1]q\in [0,1], showing that a transition from better-than-square-root cancellation to growth like a power of log⁡x\log x occurs when qq crosses 1/21/2. In particular, the first absolute moments are of order (log⁡log⁡x)3/4(\log\log x)^{3/4}. The proof proceeds by comparing these sums to integrals of random Euler products, which are then studied by analogy with Gaussian multiplicative chaos integrals. The main difficulty lies in proving the lower bound at q=1/2q=1/2, for which we also adapt arguments from the study of random walks.

Introduction

Let Xe:N→{±1}X^{e} : \mathbb{N} \to\{\pm1\} be the (random) completely multiplicative function whose values at the primes (Xe(p))p prime(X^{e}(p))_{p\ \mathrm{prime}} are independent and uniformly distributed on {±1}\{\pm1\}. This random function is sometimes referred to as an extended Rademacher random multiplicative function. We further denote by X:N→{±1,0}X : \mathbb{N} \to\{\pm1,0\} the Rademacher random multiplicative function X:=μ2XeX := \mu^{2}X^{e}, which we have chosen here to be defined by the same realisation of the values of XeX^{e} at the primes.

This paper is primarily concerned with the study of low moments (that is, the 2q2q–th moment for q∈[0,1]q \in[0,1]) of the partial sums

∑n≤xXe(n)\sum_{n\leq x} X^{e}(n)

which are meant to model the statistical properties of ∑n≤xχ(n)\sum_{n\leq x}\chi(n) for a random quadratic Dirichlet character χ\chi. To see how, choose a large parameter DD, and let dd be chosen uniformly at random among fundamental discriminants in [D,2D][D,2D]. Then the (random) quadratic character χ=χd\chi=\chi_d associated to dd is completely multiplicative, and it is straightforward to verify that the distribution of χ(p)\chi(p) for a given prime pp is approximately uniform on {±1}\{\pm1\}. One can also show that χ(p)\chi(p) and χ(p′)\chi(p') are approximately independent for different primes pp and p′p', so long as they are not too large compared to DD.

The study of random character sums is usually much more tractable than that of ∑n≤xχd(n)\sum_{n\leq x}\chi_d(n) for a fixed χd\chi_d, as it gives access to various tools which are not a priori applicable to the latter. Indeed, we are essentially only able to detect non-trivial cancellation in ∑n≤xχd(n)\sum_{n\leq x}\chi_d(n) when x≥d1/4+o(1)x \geq d^{1/4+o(1)}, using Burgess’ bound [5]. Even then, the resulting bound is much weaker than the “square-root cancellation” one might expect (meaning a bound of the shape ≪x1/2+ε\ll x^{1/2+\varepsilon}), with such a bound only being known for x≥d1−o(1)x \geq d^{1-o(1)}, as a consequence of the Pólya–Vinogradov inequality (cf. [26 Chapter 9.4]). By contrast, square-root cancellation is known to hold in the random setting: uniformly in all xx and DD, one has

1#{D<d≤2D:d fund. disc.}∑D<d≤2D∗∣1x∑n≤xχd(n)∣2≪log⁡D,(1)\frac{1}{\#\{D<d\leq2D : d\ \mathrm{fund.\ disc.}\}}\sum_{D<d\leq2D}^{*}\left|\frac{1}{\sqrt{x}}\sum_{n\leq x}\chi_d(n)\right|^{2}\ll\log D, \tag*{(1)}

where the sum ranges over fundamental discriminants [2]. Averaging over the character also allows for comparison with random models such as XX and XeX^{e}, an approach pioneered in [12]. Furthering a line of investigation initiated by Davenport and Erdős [7], recent works of Harper [19] and Nosal [Nos25] for instance used the extended model XeX^{e} to study the distribution of quadratic character sums with a random starting point. Both works rely on establishing an estimate of the form

log⁡PPη∑P≤p≤P+Pηp prime∣∑n≤Nαn(np)∣2≪E[∣∑n≤NαnXe(n)∣2]+NPη/2(∑n≤N∣αn∣)2.\frac{\log P}{P_{\eta}}\sum_{\substack{P\le p\le P+P^{\eta}\\p\ \mathrm{prime}}}\left|\sum_{n\le N}\alpha_n\left(\frac{n}{p}\right)\right|^2\ll\mathbb{E}\left[\left|\sum_{n\le N}\alpha_nX^{e}(n)\right|^2\right]+\frac{N}{P^{\eta/2}}\left(\sum_{n\le N}|\alpha_n|\right)^2.

for η∈[0.01,1]\eta\in[0.01,1] and (αn)n(\alpha_n)_n a sequence of complex numbers (see Number Theory Result/Fact 3 in [19 Nos25]).

There are several challenges related to the study of the extended Rademacher model compared to the other two main models (namely Steinhaus and Rademacher), chief among them the lack of orthogonality: whereas E[X(n)X(m)]=1(n=m)\mathbb{E}[X(n)X(m)]=\mathbf{1}(n=m) on squarefree numbers n,mn,m,

E[Xe(n)Xe(m)]=1(nm=□).\mathbb{E}[X^{e}(n)X^{e}(m)]=\mathbf{1}(nm=\square).

By factoring out the largest square divisor of nn and mm, this for instance implies the second moment estimate

E[∣1x∑n≤xXe(n)∣2]=1x∑n1,n2≤x1(n1n2=□)=1x∑a≤xμ(a)2∑n1,n2≤x/a1≍∑a≤xμ(a)2a≍log⁡x,\mathbb{E}\left[\left|\frac{1}{\sqrt{x}}\sum_{n\le x}X^{e}(n)\right|^2\right]=\frac{1}{x}\sum_{n_1,n_2\le x}\mathbf{1}(n_1n_2=\square)=\frac{1}{x}\sum_{a\le x}\mu(a)^2\sum_{\substack{n_1,n_2\le\sqrt{x/a}}}1\asymp\sum_{a\le x}\frac{\mu(a)^2}{a}\asymp\log x,

(see, e.g., [26 Theorem 2.2]). This suggests that the upper bound in (1.1) is optimal on a wide range of DD. Moreover, by Hölder’s inequality, this also implies that

E[∣1x∑n≤xXe(n)∣2q]≪(log⁡x)q,q∈[0,1].(2)\mathbb{E}\left[\left|\frac{1}{\sqrt{x}}\sum_{n\le x}X^{e}(n)\right|^{2q}\right]\ll(\log x)^q,\qquad q\in[0,1]. \tag*{(2)}

Comparing this to the statement of Theorem 1.1 shows that unlike for partial sums of XX, where the analogous “trivial” bound only loses (log⁡log⁡x)O(1)(\log\log x)^{O(1)}, the bound in (1.2) overshoots by a power of log⁡x\log x when q∈[0.01,0.99]q\in[0.01,0.99], say.

Main results

Our primary goal is to establish the following theorem, precisely determining the order of magnitude of the low moments of ∑n≤xXe(n)\sum_{n\le x}X^{e}(n).

Theorem 1.1. Let 0≤q≤10\le q\le1 be fixed. Uniformly for x≥3x\ge3, we have

E[∣1x∑n≤xXe(n)∣2q]≍{(log⁡log⁡x)−q/2,if 0≤q<12,(log⁡log⁡x)3/4,if q=12,(log⁡x)q(2q−1),if 12<q≤1.\mathbb{E}\left[\left|\frac{1}{\sqrt{x}}\sum_{n\le x}X^{e}(n)\right|^{2q}\right]\asymp\begin{cases}(\log\log x)^{-q/2},&\text{if }0\le q<\frac{1}{2},\\(\log\log x)^{3/4},&\text{if }q=\frac{1}{2},\\(\log x)^{q(2q-1)},&\text{if }\frac{1}{2}<q\le1.\end{cases}

In doing so, we also establish the analogous claim for a closely related weighted sum of XX.

Theorem 1.2. Let 0≤q≤10\le q\le1 be fixed. Uniformly for x≥3x\ge3, we have

E[∣∑n≤xX(n)n∣2q]≍{(log⁡log⁡x)−q/2,if 0≤q<12,(log⁡log⁡x)3/4,if q=12,(log⁡x)q(2q−1),if 12<q≤1.\mathbb{E}\left[\left|\sum_{n\le x}\frac{X(n)}{\sqrt{n}}\right|^{2q}\right]\asymp\begin{cases}(\log\log x)^{-q/2},&\text{if }0\le q<\frac{1}{2},\\(\log\log x)^{3/4},&\text{if }q=\frac{1}{2},\\(\log x)^{q(2q-1)},&\text{if }\frac{1}{2}<q\le1.\end{cases}

References

  1. [1][ABH17] Louis-Pierre Arguin, David Belius, and Adam J. Harper. Maxima of a randomized Riemann zeta function, and branching random walks. Ann. Appl. Probab., 27(1):178–215, 2017.arxiv.org/abs/1506.00629
  2. [2][Arm99] M. V. Armon. Averages of real character sums. J. Number Theory, 77(2):209–226, 1999.DOI
  3. [3][BM03] E. Bacry and J. F. Muzy. Log-infinitely divisible multifractal processes. Comm. Math. Phys., 236(3):449–475, 2003.
  4. [4][BRR76] R. N. Bhattacharya and R. Ranga Rao. Normal Approximation and Asymptotic Expansions. Wiley Series in Probability and Mathematical Statistics. John Wiley & Sons, New York-London-Sydney, 1976.
  5. [5][Bur63] D. A. Burgess. On character sums and L-series. II. Proc. London Math. Soc. (3), 13:524–536, 1963.DOI
  6. [6][Cai24] Rachid Caich. Random multiplicative functions and typical size of character in short intervals, 2024. https://arxiv.org/abs/2402.06426.
  7. [7][DE52] H. Davenport and P. Erdős. The distribution of quadratic and higher residues. Publ. Math. Debrecen, 2:252–265, 1952.
  8. [8]François David, Antti Kupiainen, Rémi Rhodes, and Vincent Vargas. Liouville quantum gravity on the Riemann sphere. Comm. Math. Phys., 342(3):869–907, 2016.arxiv.org/abs/1410.7318
  9. [9]Bertrand Duplantier, Rémi Rhodes, Scott Sheffield, and Vincent Vargas. Renormalization of critical Gaussian multiplicative chaos and KPZ relation. Comm. Math. Phys., 330(1):283–330, 2014.
  10. [10]Yan V. Fyodorov and Jean-Philippe Bouchaud. Freezing and extreme-value statistics in a random energy model with logarithmically correlated potential. J. Phys. A, 41(37):372001, 12, 2008.DOI
  11. [11]Maxim Gerspach. Low pseudomoments of the Riemann zeta function and its powers. International Mathematics Research Notices, 2022(1):625–664, 2022.
  12. [12]Andrew Granville and Kannan Soundararajan. Large character sums. Journal of the American Mathematical Society, 14(2):365–397, 2001.
  13. [13]Allan Gut. Probability: A Graduate Course, volume 75 of Springer Texts in Statistics. Springer, New York, 2nd edition, 2013.
  14. [14]Ofir Gorodetsky and Mo Dick Wong. A short proof of Helson’s conjecture. Bull. Lond. Math. Soc., 57(4):1065–1076, 2025.DOI
  15. [15]Jad Hamdan. Partial sums of random multiplicative functions with supercritical divisor twists, 2026. https://arxiv.org/abs/2604.05563.
  16. [16]Adam J Harper. Moments of random multiplicative functions, II: High moments. Algebra & Number Theory, 13(10):2277–2321, 2019.DOI
  17. [17]Adam J. Harper. Moments of random multiplicative functions, I: Low moments, better than squareroot cancellation, and critical multiplicative chaos. Forum of Mathematics, Pi, 8, 2020.arxiv.org/abs/1703.06654
  18. [18]Adam J Harper. Almost sure large fluctuations of random multiplicative functions. International Mathematics Research Notices, 2023(3):2095–2138, 2023.
  19. [19]Adam J. Harper. A note on character sums over short moving intervals. J. Inst. Math. Jussieu, 24(4):1395–1427, 2025.arxiv.org/abs/2203.09448
  20. [20]Adam J. Harper, Ashkan Nikeghbali, and Maksym Radziwiłł. A note on Helson’s conjecture on moments of random multiplicative functions. In Analytic Number Theory: In Honor of Helmut Maier’s 60th Birthday, pages 145–169. Springer, 2015.DOI
  21. [21]Yichao Huang, Rémi Rhodes, and Vincent Vargas. Liouville quantum gravity on the unit disk. Ann. Inst. Henri Poincaré Probab. Stat., 54(3):1694–1730, 2018.arxiv.org/abs/1502.04343
  22. [22]Adam J. Harper, Kannan Soundararajan, and Max Wenqiang Xu. Distribution of random multiplicative functions in short intervals, with proper normalization, 2026. https://arxiv.org/abs/2606.29040.
  23. [23]Antti Kupiainen, Rémi Rhodes, and Vincent Vargas. Integrability of Liouville theory: proof of the DOZZ formula. Ann. of Math. (2), 191(1):81–166, 2020.
  24. [24]J. P. Keating and N. C. Snaith. Random matrix theory and L-functions at s = 1/2. Comm. Math. Phys., 214(1):91–110, 2000.
  25. [25]Alex Kontorovich, Terence Tao, and The Prime Number Theorem+ Project. PrimeNumberTheoremAnd: A formalization of the prime number theorem and related results in Lean 4. https://github.com/AlexKontorovich/PrimeNumberTheoremAnd, 2024.
  26. [26]Hugh L. Montgomery and Robert C. Vaughan. Multiplicative Number Theory I. Classical Theory, volume 97 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2007.
  27. [28]Ellen Powell. Critical Gaussian multiplicative chaos: a review. Markov Process. Related Fields, 27(4):557–606, 2021.arxiv.org/abs/2006.13767
  28. [29]Guillaume Remy. The Fyodorov–Bouchaud formula and Liouville conformal field theory. Duke Math. J., 169(1):177–211, 2020.DOI
  29. [30][RV14] Rémi Rhodes and Vincent Vargas. Gaussian multiplicative chaos and applications: a review. Probab. Surv., 11:315–392, 2014.arxiv.org/abs/1305.6221
  30. [31][RZ20] Guillaume Remy and Tunan Zhu. The distribution of Gaussian multiplicative chaos on the unit interval. Ann. Probab., 48(2):872–915, 2020.arxiv.org/abs/1804.02942

Paper details

vlnc:W_zxkfb22c
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Claude

View statement
Looking for related papers…
Contents