Abstract:We prove that in the limit of large dimension, the distribution of the logarithm of the characteristic polynomial of a generalized Wigner matrix converges to a log-correlated field. In particular, this shows that the limiting joint fluctuations of the eigenvalues are also log-correlated. Our argument mirrors that of \cite{BouMod2019}, which is in turn based on the three-step argument of \cite{ErdPecRmSchYau2010,ErdSchYau2011Uni}, but applies to a wider class of models, and at the edge of the spectrum. We rely on (i) the results in the Gaussian cases, special cases of the results in \cite{BouModPai2021}, (ii) the local laws of \cite{ErdYauYin2012}(iii) the observable \cite{Bou2020} introduced and its analysis of the stochastic advection equation this observable satisfies, and (iv) the argument for a central limit theorem on mesoscopic scales in \cite{LanLopSos2021}. For the proof, we also establish a Wegner estimate and local law down to the microscopic scale, both at the edge of the spectrum.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the distribution characteristics of the logarithmic characteristic polynomial of generalized Wigner matrices in the high - dimensional limit. Specifically, the author proves that the logarithmic characteristic polynomial of generalized Wigner matrices converges to a log - correlated field in the large - dimension limit. This means that in this limit, the joint fluctuations of eigenvalues are also logarithmically correlated.
### Background and Motivation
1. **Importance of the Logarithmic Characteristic Polynomial**
- The logarithmic characteristic polynomial is of great significance in random matrix theory, especially when studying the fluctuation properties of eigenvalues.
- For example, for Haar - distributed unitary matrices (i.e., the circular unitary ensemble, CUE), the logarithmic characteristic polynomial can be used as a model for the Riemann ζ - function on the critical axis, and a central limit theorem similar to Selberg has been proven.
2. **Existing Research**
- Many previous studies have proven the distribution characteristics of the logarithmic characteristic polynomial in different random matrix models. For example, for GUE (Gaussian unitary ensemble), the logarithmic characteristic polynomial converges in the distribution sense to a random Fourier series.
- In the point - wise sense, some central limit theorems have also proven the behavior of the logarithmic characteristic polynomial at specific points.
### Research Objectives
- **Main Results**
- The author proves that the logarithmic characteristic polynomial of generalized Wigner matrices is log - correlated in the large - dimension limit.
- Specifically, for a fixed small quantity ε and a constant c, define a subset \( G_{\epsilon, c} \) of the spectrum, and consider multiple points \( E_1, E_2,\ldots, E_m \) in this subset. The author proves that after appropriate standardization, the joint distribution of the real and imaginary parts of the logarithmic characteristic polynomial at these points converges to a multivariate normal distribution, and the covariance matrix has a specific form.
- **Methods**
- The author uses a three - step strategy, which is based on the results of Wegner estimates and local laws on the optimal scale.
- By transforming the problem into a problem of fluctuations of linear statistics on a near - macroscopic scale, the author is able to use the existing central limit theorems to complete the proof.
### Technical Details
- **Wegner Estimates**
- Wegner estimates show that in any arbitrarily small interval, it can be almost surely determined that no eigenvalues exist. This is especially important in the edge spectral region.
- **Local Laws**
- Local laws describe the closeness between the Green's function of a matrix and the corresponding mean - field theory in the large - dimension limit.
- **Central Limit Theorem**
- The author also proves a central limit theorem on a near - macroscopic scale, which is crucial for controlling the behavior of the characteristic polynomial.
### Conclusions
- The results of this paper not only expand the existing understanding of the logarithmic characteristic polynomial in random matrix theory, but also provide new tools and methods for studying the eigenvalue fluctuations of generalized Wigner matrices.
### Application Prospects
- These results have potential applications in statistical physics, quantum chaos, number theory, etc., especially when studying the statistical properties of complex systems.
Through these detailed technical analyses, the author successfully proves that the logarithmic characteristic polynomial of generalized Wigner matrices is log - correlated in the large - dimension limit, thus providing an important theoretical basis for research in this field.