Skip to content
Preprint

The Berry--Esseen Estimate in the Free Central Limit Theorem

Aug 2026 · 0 citations · 9 references
Mathematics

Abstract

We consider sums of freely independent self-adjoint random variables that are not necessarily identically distributed. Let $\mu_j$ denote the distribution of the $j$th summand. We assume that they have mean zero and finite absolute moments of order $2+\delta$, where $0<\delta\le 1$. Let $\Delta$ denote the Kolmogorov distance, let $\mu^{(n)}$ be the distribution of the normalized partial sum, let $\omega$ be the standard semicircle law, and let $B_n^2$ be the variance of the partial sum. The purpose of this paper is to prove the Berry--Esseen estimate in the free central limit theorem. Namely, there exists an absolute constant $C>0$ such that, for every $0<\delta\le 1$, \[ \Delta(\mu^{(n)},\omega) \le \frac{C}{B_n^{2+\delta}}\sum_{j=1}^n \int_{\R}|x|^{2+\delta}\,\mu_j(dx), \] Our result not only improves several known estimates for general non-identically distributed random variables, but also establishes exactly the same Berry--Esseen estimate as in classical probability theory. The proof combines truncation, a quantitative estimate for the $R$-transform, a stability analysis of a perturbed semicircle equation, and a Bai-type smoothing inequality.

View source

Similar papers

Preprint Aug 2026

The Optimal Rate in the Averaged Random-Marginal Central Limit Theorem for Log-Concave Measures

Let $X$ be a centered isotropic log-concave random vector in $\mathbb{R}^n$. For $\theta\in S^{n-1}$, let $\mu_\theta$ be the law of $\langle X,\theta\rangle$, and let $\Theta$ be uniformly distributed on $S^{n-1}$, independently of $X$. We prove the sharp estimate \[ \textsf{E} W_1(\mu_\Theta,\gamma_1) \le \frac{C}{n}. \] Here the Wasserstein distance is computed after the direction is fixed and is then averaged over the sphere. No symmetry assumption is imposed. A product measure with centered exponential coordinates gives a matching lower bound of order $n^{-1}$. The proof separates the averaged-direction law from the fluctuation among fixed directions. For the first part, a Taylor expansion in the random radius retains a mean-zero cancellation and yields an $O(n^{-1})$ error. For the second, a weighted $L^2$ distance between distribution functions is converted into an exact spherical kernel depending only on $|x|^2$, $|y|^2$, and $\langle x,y\rangle$. Expanding this kernel in $\langle x,y\rangle$, we control its linear, quadratic, and cubic terms using the quadratic variance inequality $\operatorname{Var}(X^\top M X) \le 8\operatorname{Tr}(M^2)$, while fixed-order moment estimates control the remainder.

Xuanang Hu · 0 citations
Preprint Aug 2026

An Exponential Lower Bound for the Permanent of Random Bernoulli Matrix

Let $M_n$ be an $n\times n$ matrix with independent uniform sign entries. We prove that there exist absolute constants $C,c>0$ such that, for all sufficiently large $n$, \[ \mathbb{P}\!\left( \left|\operatorname{Per}(M_n)\right| \ge e^{-Cn}\sqrt{n!} \right) \ge 1-n^{-c}. \] Our proof tracks the total squared permanent of minors under successive row exposure. Up to $k=\lfloor n/2\rfloor$, the total squared permanent grows deterministically via the Boolean lattice up-operator; for larger $k$, the row exposure increments are governed by positive semidefinite Rademacher quadratic forms. Therefore, we confirms the exponential scale lower bound suggested by Tao and Vu.

Yiming Chen · 0 citations
Preprint Jul 2026

Asymptotically sharp Hardy-Rellich inequalities on lattices

We determine the sharp asymptotic behavior of the optimal constants in the discrete Hardy-Rellich inequalities on the lattice $\mathbb{Z}^d$. For every fixed integer $m\ge 1$, let ${\mathcal C}_{m,d}$ be the best constant in the $m$-th order Hardy-Rellich inequality. We prove that $$\lim_{d\to\infty}\frac{\mathcal C_{m,d}}{d^m}=2^m.$$ Our approach combines a Fourier reduction to weighted inequalities with the flat torus and general weighted Hardy-Rellich identities of first and second order. A key novelty is the use of probabilistic concentration estimates, specifically Hoeffding's inequality and entropy methods, to handle estimates involving the anisotropic weight $\omega^\gamma$ $(\gamma\geq 1)$ where $$\omega(x)=\sum_{j=1}^{d} \Big(\sin\frac{x_j}{2}\Big)^2$$ in a dimension-uniform manner. These tools yield asymptotically sharp weighted estimates on the torus, from which the lattice inequalities follow by iteration.

Xia Huang, Dong Ye · 1 citation
Preprint Jul 2026

Exterior power sums

We prove that for every fixed $\lambda>0$ and all sufficiently large $n$, any $z_1,\dots,z_n\in\C$ with $|z_j|\geq1$ satisfy $\max_{2\leq k\leq n+1}|\sum_j z_j^k|>e^{-\lambda n}$. Consequently, the $n$th root of the optimal maximum tends to $1$, so no constant $C>1$ in Erd\H{o}s 973 can exist. The proof combines a truncated exponential factorization with overconvergence on an open set outside the unit disk and a normal-family obstruction for Cauchy transforms.

Yanping Luo, Ruiyi Yang, Keheng Zhu · 1 citation
Preprint Aug 2026

Power-sum convergence for randomly weighted means and Breiman's conjecture

We establish a power-sum convergence principle for randomly weighted means. Let $P(t)=(P_j(t))$ be random finitely supported subprobability weight sequences, independent of iid centered integrable marks. If the expected total mass converges and the expected power sum of some order $r\in(0,1)$ is uniformly bounded, then convergence, for one fixed mark law, of $\sum_jP_j(t)X_j$ to a nondegenerate law forces $\mathbb{E}\sum_jP_j(t)^p$ to converge to a positive limit for some $p\in(1,2]$. The proof combines normal-family compactness for Mellin transforms of characteristic-function remainders, inversion at frequencies $\pm u$, and Landau's theorem at the abscissa of convergence; a uniform Abelian estimate handles the endpoint $p=2$. For weights obtained by normalizing a Poisson-sized iid sample of nonnegative variables, a gap below one for the logarithmic slope of the Laplace exponent yields the required power-sum bound of order below one. A Poissonized ratio Tauberian theorem then identifies the common tail of the unnormalized variables as regularly varying, with index $-\beta$ for a unique $\beta\in[0,1)$. As an application, this proves the remaining necessity direction in Breiman's 1965 conjecture for centered integrable marks. Combined with Breiman's sufficiency theorem, it completes the conjecture.

J. Lenzi · 0 citations
Preprint Aug 2026

Central limit theorem in R\'enyi divergence for lattice random variables

We establish a central limit theorem in R\'enyi divergence for independent and identically distributed lattice random variables $X_1, \cdots, X_n$ with zero mean, unit variance, and maximal span $h>0$. Let $S_n=(X_1+\cdots+X_n)/\sqrt n$. Let $Z_n$ denote the standard Gaussian distribution quantized on the support lattice of $S_n$. For every $\alpha>1$, with $\beta=\alpha/(\alpha-1)$, we prove that the R\'enyi divergence $D_\alpha(S_n\|Z_n)\to 0$ if and only if the divergence is finite at some convolution level and the strict sub-Gaussian condition $$ \mathbb E e^{tX}<e^{\beta t^2/2},\quad t\in\mathbb R,~ t\ne0 $$ holds. Under these conditions, we further derive an Edgeworth-type asymptotic expansion of the divergence to arbitrary order. These results provide a lattice counterpart of the R\'enyi entropic central limit theorem for continuous random variables due to Bobkov, Chisyakov and G\"{o}tze (\emph{Ann. Probab.} \textbf{47} (2019), 270--323).

Zhen Fu, Jiange Li · 0 citations