Abstract
Let \(z\in \mathbb {C}\), let \(\sigma ^2>0\) be a variance, and for \(N\in \mathbb {N}\) define the integrals
These are expected values of the polynomials \(P_N^{}(z)=\prod _{1\le n\le N}(X_n^2+z^2)\) whose 2N zeros \(\{\pm i X_k\}^{}_{k=1,\ldots ,N}\) are generated by N identically distributed multi-variate mean-zero normal random variables \(\{X_k\}^{N}_{k=1}\) with co-variance \(\mathrm{{Cov}}_N^{}(X_k,X_l)=(1+\frac{\sigma ^2-1}{N})\delta _{k,l}+\frac{\sigma ^2-1}{N}(1-\delta _{k,l})\). The \(E_N^{}(z;\sigma )\) are polynomials in \(z^2\), explicitly computable for arbitrary N, yet a list of the first three \(E_N^{}(z;\sigma )\) shows that the expressions become unwieldy already for moderate N—unless \(\sigma = 1\), in which case \(E_N^{}(z;1) = (1+z^2)^N\) for all \(z\in \mathbb {C}\) and \(N\in \mathbb {N}\). (Incidentally, commonly available computer algebra evaluates the integrals \(E_N^{}(z;\sigma )\) only for N up to a dozen, due to memory constraints). Asymptotic evaluations are needed for the large-N regime. For general complex z these have traditionally been limited to analytic expansion techniques; several rigorous results are proved for complex z near 0. Yet if \(z\in \mathbb {R}\) one can also compute this “infinite-degree” limit with the help of the familiar relative entropy principle for probability measures; a rigorous proof of this fact is supplied. Computer algebra-generated evidence is presented in support of a conjecture that a generalization of the relative entropy principle to signed or complex measures governs the \(N\rightarrow \infty \) asymptotics of the regime \(iz\in \mathbb {R}\). Potential generalizations, in particular to point vortex ensembles and the prescribed Gauss curvature problem, and to random matrix ensembles, are emphasized.
Similar content being viewed by others
Notes
Kirchhoff’s Hamilton function generates point vortex motion in \(\mathbb {R}^2\) without boundaries or externally produced stream functions. The a-priori measure \(\mu _0\) adds an external stream function \(\sum _k\ln f(\mathbf {s}_k)\) to \(-{\scriptstyle {\frac{\beta }{N}}}H^{(N)}\) to prevent the vortices from escaping to spatial \(\infty \).
Louis Nirenberg originally posed the problem for the prescribed Gauss curvature equation on the sphere \(\mathbb {S}^2\), which is much harder to answer due to some topological obstructions (see, e.g., [1, 12, 15]). One such obstruction translates into the interesting requirement that the (rescaled) reciprocal Onsager temperature \(\beta =-8\pi \), but this is exactly the borderline value where the canonical ensemble becomes a singular measure and therefore fails to supply partial answers to Nirenberg’s question. However, as explained in [17, 20], the microcanonical point vortex ensemble [25] will produce solutions to the prescribed Gauss curvature equation on \(\mathbb {S}^2\) whenever some exist, although only maximum entropy solutions can be produced.
Personal communication from A.C. to M.K.; ca. 2000.
Probabilists prefer the opposite sign convention for the relative entropy; cf. [9].
Estimates in the opposite direction follow from \((x_k-x_l)^2\ge 0\) with “\(=\)” iff \(x_k=x_l\), thus
$$\begin{aligned} \sigma E_N^{}(z;\sigma )\; \left\{ \begin{array}{lll} &{}> \sigma ^{N}(\sigma ^2+z^2)^N \quad \text{ if }\quad \sigma<1\\ &{} < \sigma ^{N}(\sigma ^2+z^2)^N \quad \text{ if }\quad \sigma >1\\ \end{array} \right. . \end{aligned}$$(62)Here: directional derivatives in the direction of any \(\psi =\psi _++\psi _-\), with \(\psi _\pm \in C^\infty \) compactly supported inside the support of the positive/negative part of the a-priori measure, respectively, and with \(\psi \) integrating to zero to preserve the normalization of \(\dot{\varsigma }^{(N)}\).
Incidentally, our study shows that the multivariate normal random variables with \(\sigma \ne 1\) will not be amongst the laws asked for in Q1.
References
Chang, S.Y.A., Yang, P.: Prescribing Gaussian curvature on \({\mathbb{S}}^2\). Acta Math. 159, 215–259 (1987)
Conrey, B.: The Riemann Hypothesis, Notices of the AMS, pp. 341–353 (2003)
Caglioti, E., Lions, P.L., Marchioro, C., Pulvirenti, M.: A special class of stationary flows for two-dimensional Euler equations: a statistical mechanics description. Commun. Math. Phys. 143, 501–525 (1992)
Caglioti, E., Lions, P.L., Marchioro, C., Pulvirenti, M.: A special class of stationary flows for two-dimensional Euler equations: a statistical mechanics description. II. Commun. Math. Phys. 174, 229–260 (1995)
Chanillo, S., Kiessling, M.K.-H.: Surfaces with prescribed Gauss curvature. Duke Math. J. 105, 309–353 (2000)
Dirac, P.A.M.: Bakerian lecture. The physical interpretation of quantum mechanics. Proc. R. Soc. A 180, 1–40 (1942)
Dumitriu, I., Edelman, A.: Matrix models for \(\beta \) ensembles. J. Math. Phys. 43, 5830–5847 (2002)
Edwards, H.M.: Riemann’s Zeta Function. Dover, New York (1974)
Ellis, R.S.: Entropy, Large Deviations, and Statistical Mechanics. Springer, New York (1985)
Forrester, P.J.: Log-Gases and Random Matrices. Princeton University Press, Princeton (2010)
Gibbs, J.W.: Elementary Principles of Statistical Mechanics. Scribner’s Sons, New York (1902)
Han, Z.C.: Prescribing Gauss curvature on \({\mathbb{S}}^2\). Duke Math. J. 61, 679–703 (1990)
Hewitt, E., Savage, L.J.: Symmetric measures on Cartesian products. Trans. Am. Math. Soc. 80, 470–501 (1955)
Iwaniec, H.: Lectures on the Riemann Zeta Function. AMS, Providence (2014)
Kazdan, J., Warner, F.: Curvature functions for compact 2-manifolds. Ann. Math. 99, 14–47 (1974)
Kiessling, M.K.-H.: Statistical mechanics of classical particles with logarithmic interactions. Commun. Pure Appl. Math. 46, 27–56 (1993)
Kiessling, M.K.-H.: Statistical mechanics approach to some problems in conformal geometry. Physica A 279, 353–368 (2000)
Kiessling, M.K.-H., Lebowitz, J.L.: The micro-canonical point vortex ensemble: beyond equivalence. Lett. Math. Phys. 42, 43–58 (1997)
Kiessling, M.K.-H., Spohn, H.: A note on the eigenvalue density of random matrices. Commun. Math. Phys. 199, 683–695 (1999)
Kiessling, M.K.-H., Wang, Y.: Onsager’s ensemble for point vortices with random circulations on the sphere. J. Stat. Phys. 148, 896–932 (2012)
Maalaoui, A.: Prescribing the Q-curvature on the sphere with conical singularities. Discret. Contin. Dyn. Syst. 36, 6307–6330 (2016)
Mehta, M.L.: Random Matrices, 2nd edn. Academic Press, New York (1991)
Messer, J., Spohn, H.: Statistical mechanics of the isothermal Lane-Emden equation. J. Stat. Phys. 29, 561–578 (1982)
Mezard, M., Parisi, G., Virasoro, M.A.: Spin Glass Theory and Beyond. World Scientific, Singapore (1987)
Onsager, L.: Statistical hydrodynamics. Nuovo Cim. Suppl. 6, 279–287 (1949)
Robinson, D.W., Ruelle, D.: Mean entropy of states in classical statistical mechanics. Commun. Math. Phys. 5, 288–300 (1967)
Shamis, M., Zeitouni, O.: The Curie–Weiss Model with Complex Temperature: Phase Transitions. arXiv:1701.02375v1 (2017)
Acknowledgements
I am grateful to Alice Chang for her question about the connection between statistical mechanics of point vortices and Nirenberg’s problem with signed Gaussian curvature, which started this inquiry. I also thank Roger Nussbaum, Alex Kontorovich, and Shadi Tahvildar-Zadeh for patiently listening to my reports of progress which helped me obtaining greater clarity in this writeup. Finally I thank both referees for their constructive criticisms.
Author information
Authors and Affiliations
Corresponding author
Additional information
Dedicated to the memory of my mother, Elisabeth Kiessling, née Appeltrath.
Appendix A: Alternate Proof of Corollary 2.3
Appendix A: Alternate Proof of Corollary 2.3
Proof of Corollary 2.3
By the permutation symmetry in \(\{x_1,\ldots ,x_N\}\) of the pdf given by the integrand of (37), for the pertinent random variables \(\{X_1,\ldots ,X_N\}\) we have
and for these (as for any) mean-zero random variables \(\{X_1,\ldots ,X_N\}\), we have
Under a Euclidean transformation (here: rotation) from \(\{x_1,\ldots ,x_N\}\) to \(\{y_1,\ldots ,y_N\}\),
By the proof of Proposition 2.1,
The proof is complete. \(\square \)
Remark 8.1
There is no unique rotation in \(\mathbb {R}^N\) which maps \(\{x_1,x_2,\ldots ,x_N\}\) into \(y_1:={\textstyle {\frac{1}{\sqrt{N}}}}\sum _{1\le k \le N} x_k\); even after stipulating that the \(x_1\) axis should be rotated into the \(y_1\) axis we are left to choose an arbitrary \(SO(N-1)\) rotation “about the \(y_1\)-axis” to fix the remaining \(N-1\) axes. Such a freedom may be useful to simplify the expected value integrals in the y coordinates, but we have not explored this here.
1.1 Appendix B: Alternate Proof of the Positivity of \(E_{2K}^{}(z;\sigma );\; K\in \mathbb {N}\), \(\sigma \ge 1\)
Proposition 8.2
Let \(K\in \mathbb {N}\). Then for \(\sigma \ge 1\) and \(z^2\in \mathbb {R}\) we have \(E_{2K}(z;\sigma )\ge 0\), with “\(=0\)” iff \(\sigma =1\) and \(z^2=-1\).
Proof of Proposition 8.2
We split the \(N=2K\) variables into two disjoint sets of size K, keeping the notation \(x_n^{}\) for \(n=1,\ldots ,K\) and renaming \(x_n^{}=:\tilde{x}_k^{}\) if \(n=K+k\) with \(k=1,\ldots ,K\). Note that
and further that
Again recalling that \((x_1^{},\ldots ,x_K^{})=:\vec {x}\), and defining
and
we have, for \(K\in \mathbb {N}\),
which, since \(\sigma \ge 1\), is manifestly \(\ge 0\). More precisely, r.h.s. (143) is estimated from below by the \(j=0\) contribution, evaluating to \(\frac{1}{\sigma }\Big [\sqrt{2\frac{\sigma ^2}{\sigma ^2+1} }(z^2 +2\frac{\sigma ^2}{\sigma ^2+1} )\Big ]^{2K}\ge 0\), with “\(=0\)” iff \(\sigma =1\) and \(z^2=-1\). When \(\sigma =1\), the \(j=0\) term is also the only contribution to r.h.s. (143). \(\square \)
Rights and permissions
About this article
Cite this article
Kiessling, M.KH. Heuristic Relative Entropy Principles with Complex Measures: Large-Degree Asymptotics of a Family of Multi-variate Normal Random Polynomials. J Stat Phys 169, 63–106 (2017). https://doi.org/10.1007/s10955-017-1843-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10955-017-1843-6