Derivative relations for determinants, Pfaffians and characteristic polynomials in random matrix theory

Abstract: Explicit expressions are proven for derivatives of the ratio of a determinant or Pfaffian determinant and a Vandermonde determinant. Such ratios appear for example in general group integrals of Harish-Chandra--Itzykson--Zuber type and in expectation values of products of characteristic polynomials in random matrix theory. In the latter case we start from known results for general non-Hermitian and Hermitian ensembles for expectation values without derivatives, at finite matrix size. They are given in terms of the determinant or Pfaffian of the corresponding kernel, for unitary or orthogonal and symplectic ensembles, respectively. Several equivalent expressions are proven for general ratios of determinants, starting from first order derivatives containing the Borel transform of the corresponding matrix or kernel. Higher order derivatives are expressed as sums over partitions containing determinants of derivatives of these, with coefficients given in terms of combinatorial expressions. Our most general result is valid for mixed higher order derivatives of ratios of determinants in several variables. This generalises previous findings, e.g. for mixed moments in specific ensembles of random matrices, relevant in applications to the Riemann $ζ$-function. Applications of our results to several examples are presented, including the complex Ginibre ensemble and the circular unitary ensemble.

• The paper presents analytic and combinatorial frameworks for expressing derivatives of determinant and Pfaffian ratios in random matrix theory.
• It introduces explicit formulas using Borel transforms and symmetric function techniques applicable to unitary, orthogonal, and symplectic ensembles.
• The research enables precise finite-N evaluations of mixed moments, linking characteristic polynomial derivatives to L-function zeros, quantum chaos, and QCD spectra.

Derivative Structures for Determinant Ratios and Characteristic Polynomials in Random Matrix Theory

Overview and Context

The paper "Derivative relations for determinants, Pfaffians and characteristic polynomials in random matrix theory" (2603.29510) develops a comprehensive framework for describing and computing derivatives of ratios of determinants and Pfaffians, particularly those appearing in expectation values of characteristic polynomials in random matrix theory (RMT). Central to its analysis are ensembles with unitary, orthogonal, or symplectic symmetry, often formulated in terms of determinantal or Pfaffian point processes. The research provides both analytic and combinatorial tools to express general derivative relations, enabling a systematic approach to correlation functions, mixed moments, and higher-order statistics connected to characteristic polynomials.

This work targets computations that arise in both mathematical and physical contexts, notably the statistics of zeros of $L$-functions (such as the Riemann $\zeta$-function), analysis of quantum chaotic systems, and Dirac operator spectrum studies in QCD. It unifies and extends prior results on characteristic polynomials by providing general, symmetry-agnostic formulas valid at finite matrix dimension, circumventing traditional complications caused by the presence of Vandermonde determinants in denominators.

Main Results: Analytic and Combinatorial Structures

The fundamental achievement is the explicit characterization of derivatives acting on general ratios of determinants (or Pfaffians) and Vandermonde determinants. These ratios typify expectation values for products and derivatives of characteristic polynomials. The authors present three threads of results:

1. General Analytic Relations: For mixed derivatives—potentially of different orders in various variables—acting on ratios of determinants (or Pfaffians) to Vandermonde determinants, the paper provides formulas expressing these as differential operators acting on Pfaffians/determinants of matrix elements, which are subjected to analytic (Borel-type or Cauchy-type) transforms. These results (Theorem 1) generalize known determinantal identities and are valid at finite $N$ without restriction to specific ensembles.
2. Specialized Formulas via Borel Transforms: For the important case of first-order derivatives (e.g., computing mixed moments involving determinants and their first derivatives), the analytic transforms become explicit Borel transforms. The derivatives can be written as polynomials in differential operators evaluated at the origin, with determinantal or Pfaffian structures in terms of these transform entries. These formulas subsume and clarify previous partial results for derivative moments of unitary characteristic polynomials.
3. Combinatorial Formulations with Kostka Numbers: The series provides a combinatorial route exploiting Schur polynomial expansions and representation-theoretical tools to give determinant expressions with coefficients involving Kostka numbers—the multiplicities of semi-standard Young tableaux with fixed shape and weight. Particularly, for higher-order derivatives, the paper connects expectation values to explicit combinatorial sums encoding the symmetric function theory underlying the determinantal/Pfaffian averages.

These analytic and combinatorial routes produce equivalent results but leverage different mathematical structures, giving flexibility for applications and the potential for further generalization.

Examples and Explicit Applications

The paper illustrates the general method by explicit calculation in canonical ensembles:

• Complex Ginibre Ensemble (non-Hermitian, unitary symmetry):

The limiting kernel structure and its derivatives are computed; all mixed moments of derivatives of characteristic polynomials simplify to combinations of Laguerre polynomials, Schur polynomials, and analytic functions, with the determinantal structure allowing compact closed-form expressions. For single- and double-derivative cases, explicit forms with clear combinatorial or orthogonal polynomial content are provided.

• Circular Unitary Ensemble (CUE):

The approach recovers, clarifies, and extends known results about joint moments of characteristic polynomials and their derivatives in CUE. In the scaling limit, the determinant/Pfaffian structure persists and links directly to known asymptotics for logarithmic derivatives of the Riemann zeta function, via Keating–Snaith-type correspondences. The explicit connection to Borel and higher-order Borel transforms, and their action on truncated geometric series kernels, enables analytic continuation and exact evaluations.

• Symplectic and Orthogonal Ensembles:

For ensembles governed by Pfaffian point processes (S, O symmetry classes), the most general results carry over, including when only pairs of eigenvalues (complex conjugate pairs or real/complex splits in orthogonal ensembles) are present.

Key Structural Claims and Theoretical Implications

• Universality and Symmetry Independence:

The results hold for general weight functions and kernel structures associated to the symmetry class, not just for the classical Gaussian/Laguerre/Hermite cases. They provide formulas, valid for all finite $N$, that can be used to take scaling limits and prove universality theorems under broad conditions.

• Remediation of Vandermonde Structure:

The technical device of replacing the reciprocal Vandermonde determinant by analytic transforms (and exploiting the de Bruijn–Andreief identities and symmetric function expansions) resolves issues with differentiating non-polynomial denominator structures, producing systematically polynomial (or holomorphic) outputs for moments and their derivatives.

• Explicit Connections to Representation Theory:

By expressing the results in terms of Schur functions, shifted partitions, and Kostka numbers, the work provides new avenues to interpret mixed derivative/correlation structures in terms of symmetric group representations, allowing combinatorial enumeration and supporting analytic continuation.

• Potential for Analytic Continuation:

Several formulas furnish analytic continuations in the total power/degree parameter (e.g., in the moment order $k$) for expectation values, generalizing previously known results in CUE and Ginibre settings.

Implications and Outlook

The analytic, algebraic, and combinatorial frameworks presented have broad implications:

• $L$-Function and Zeta Function Statistics:

The methodologies enable rigorous asymptotic and finite-$N$ analysis of joint moments involving derivatives of characteristic polynomials, directly relevant to the modeling of moments and correlations of the Riemann zeta function and its analytic analogs.

• Quantum Chaology and Periodic Orbit Theory:

The ability to compute mixed moments of secular determinants and derivatives as polynomials in eigenvalues underpins periodic orbit expansions in quantum chaotic systems, with potential utility for spectral statistics and universality investigations.

• Large-$N$ Limit, Universality, and New Scaling Regimes:

The symmetry-class independence and explicit finite-$N$ structure facilitate the derivation of scaling limits, furthering both universality arguments and precise analysis at spectral 'edges' or other critical regimes.

• Extension to Ratio Averages and Cauchy Structure:

While the focus is on product averages and their derivatives, the authors note that their analytic machinery extends (with further development) to ratios of characteristic polynomials, relevant to more refined observables and recently in use for topological/field-theoretical random matrix problems.

Conclusion

This work establishes a versatile and technically robust toolkit for computing and interpreting derivatives of ratios of determinantal and Pfaffian structures, with direct implications for random matrix theory, number theory, and mathematical physics. The connection to symmetric function theory and combinatorics further broadens the class of analytic tools available for the study of characteristic polynomials. This technical systematization is expected to have lasting impact on the analysis of complex correlation functions and mixed moment problems, and opens avenues for further developments in ratio statistics and new universality classes in RMT.