# The Real Ginibre Ensemble with $$k=O(n)$$ Real Eigenvalues

• Published:

## Abstract

We consider the ensemble of real Ginibre matrices conditioned to have positive fraction $$\alpha >0$$ of real eigenvalues. We demonstrate a large deviations principle for the joint eigenvalue density of such matrices and introduce a two phase log-gas whose stationary distribution coincides with the spectral measure of the ensemble. Using these tools we provide an asymptotic expansion for the probability $$p^n_{\alpha n}$$ that an $$n\times n$$ Ginibre matrix has $$k=\alpha n$$ real eigenvalues and we characterize the spectral measures of these matrices.

This is a preview of subscription content, log in via an institution to check access.

## Subscribe and save

Springer+ Basic
\$34.99 /Month
• Get 10 units per month
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

## Notes

1. For the sake of completeness, we recall that the distance between two probability measures $$(\mu ,\nu )$$ on $$\mathbbm {C}$$ is defined as:

\begin{aligned} d_L(\mu ,\nu )=\inf \left\{ \delta >0\,;\,\mu (A)\le \nu (A^\delta ) \text{ and } \nu (A)\le \mu (A^\delta )\quad \forall A\in \mathcal {B}(\mathbb {C})\right\} \end{aligned}

where $$\mathcal {B}(\mathbb {C}$$ is the Borel algebra of $$\mathbbm {C}$$ and for $$A\in \mathcal {B}(\mathbbm {C})$$, $$A^\delta =\{x;d(x,A)\le \delta \}$$.

## References

1. Forrester, P.J.: London Mathematical Society Monographs. Princeton University Press, Princeton (2010)

2. Tao, T.: Topics in Ramdom Matrix Theory. American Mathematical Society, Providence (2012)

3. Bordenave, C., Chafai, D.: Around the circular law. Probab. Surv. 93, 1–89 (2012)

4. Wigner, E.P.: Lower limit for the energy derivative of the scattering phase shift. Phys. Rev. 98(1), 145–147 (1955)

5. Auffinger, A., Ben Arous, G., Černỳ, J.: Random matrices and complexity of spin glasses. Commun. Pure Appl. Math. 66(2), 165–201 (2013)

6. May, R.M.: Will a large complex system be stable? Nature 238, 413–414 (1972)

7. Wainrib, G., Touboul, J.: Topological and dynamical complexity of random neural networks. Phys. Rev. Lett. 110(11), 118101 (2013)

8. del Molino, L.C.G., Pakdaman, K., Touboul, J., Wainrib, G.: Synchronization in random balanced networks. Phys. Rev. E 88(4), 042824 (2013)

9. Couillet, R., Debbah, M., et al.: Random Matrix Methods for Wireless Communications. Cambridge University Press, Cambridge (2011)

10. Jaeger, H., Haas, H.: Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication. Science 304(5667), 78–80 (2004)

11. Donoho, D.L.: Compressed sensing. IEEE Trans. Inf. Theory 52(4), 1289–1306 (2006)

12. Lehmann, N., Sommers, H.J.: Eigenvalue statistics of random real matrices. Phys. Rev. Lett. 67, 941–944 (1991)

13. Edelman, A.: The probability that a random real Gaussian matrix has k real Eigenvalues, related distributions, and the circular law. J. Multivar. Anal. 60, 203–232 (1997)

14. Forrester, P.J., Nagao, T.: Eigenvalue statistics of the real Ginibre enesmble. Phys. Rev. Lett. 99, 050603 (2007)

15. Sommers, H.J.: Symplectic structure of the real Ginibre ensemble. J. Phys. A 40(29), F671 (2007)

16. Forrester, P.J., Nagao, T.: Skew orthogonal polynomials and the partly symmetric real Ginibre ensemble. J. Phys. A 41(37), 375003 (2008)

17. Forrester, P.J., Mays, A.: A method to calculate correlation functions for $$\beta =1$$ random matrices of odd size. J. Phys. A 134(3), 443–462 (2009)

18. Sinclair, C.: Correlation functions for $$\beta =1$$ ensembles of matrices of odd size. J. Stat. Phys. 136(1), 17–33 (2009)

19. Sommers, H.J., Wieczorek, W.: General eigenvalue correlations for the real Ginibre ensemble. J. Phys. A 41(40), 405003 (2008)

20. Borodin, A., Sinclair, C.: The Ginibre ensemble of real random matrices and its scaling limits. Commun. Math. Phys. 291(1), 177–224 (2009)

21. Rider, B., Sinclair, C.D., et al.: Extremal laws for the real ginibre ensemble. Ann. Appl. Probab. 24(4), 1621–1651 (2014)

22. Tao, T., Vu, V.: Random matrices: universality of ESD and the circular law. Ann. Probab. 38(5), 2023–2065 (2010)

23. Tao, T., Vu, V.: Random matrices: universality of local spectral statistics of non-hermitian matrices (2012). arXiv:1206.1893

24. Bourgade, P., Erdos, L., Yau, H.: Universality of general $$beta$$-ensembles. Duke Mathematical Journal 163(6), 1127–1190 (2014)

25. Bourgade P., Erdos, L., Yau, H., Yin, J.: Fixed energy universality for generalized Wigner matrices (2014). arXiv:1407.5606

26. Erdös, L., Yau, H.-T., Yin, J.: Rigidity of eigenvalues of generalized wigner matrices. Adv. Math. 229(3), 1435–1515 (2012)

27. Bourgade, P., Erdös, L., Yau, H.-T.: Bulk universality of general $$\beta$$-ensembles with non-convex potential. J. Math. Phys. 53(9), 095221 (2012)

28. Bourgade, P., Erdös, L., Yau, H.-T.: Edge universality of beta ensembles. Commun. Math. Phys. 332(1), 261–353 (2014)

29. Bourgade, P., Yau, H.-T., Yin, J.: Local circular law for random matrices. Probab. Theory Relat. Fields 159(3–4), 545–595 (2014)

30. Bourgade, P., Yau, H.-T., Yin, J.: The local circular law ii: the edge case. Probab. Theory Relat. Fields 159(3–4), 619–660 (2014)

31. Ben Arous, G., Guionnet, A.: Large deviations for Wigner’s law and Voiculescu’s non-commutative entropy. Probab. Theory Relat. Fields 108, 517–542 (1997)

32. Ben Arous, G., Zeitouni, O.: Large Deviations from the circular law. ESAIM: Probab. Stat. 2, 123–134 (1998)

33. Edelman, A., Kostlan, E., Shub, M.: How many Eigenvalues of a random matrix are real? J. Am. Math. Soc. 7, 247–267 (1994)

34. Ginibre, J.: Statistical ensembles of complex, quaternion, and real matrices. J. Math. Phys. 6, 440–449 (1965)

35. Sommers, H.J., Crisanti, A., Sompolinsky, H.: Spectrum of large random asymmetric matrices. Phys. Rev. Lett. 60, 1895 (1988)

36. Tribe, R., Zaboronski, O.: Pfaffian formulae for one dimensional coalescing and annihilating systems. Electron. J. Probab. 163(76), 2080–2103 (2011)

37. Forrester, P.J.: Diffusion processes and the asymptotic bulk gap probability for the real Ginibre ensemble. arXiv:1306.4106

38. Beenakker, C.: Random-matrix theory of majorana fermions and topological superconductors (2014). arXiv:1407.2131

39. Kanzieper, E., Akemann, G.: Statistics of real eigenvalues in Ginibre’s ensemble of random real matrices. Phys. Rev. Lett. 95(23), 230201 (2005)

40. Akemann, G., Kanzieper, E.: Integrable structure of Ginibre’s ensemble of real random matrices and a Pfaffian integration theorem. J. Stat. Phys. 129(5–6), 1159–1231 (2007)

41. Kanzieper, E., Poplavskyi, M., Timm, C., Tribe, R., Zaboronski, O.: What is the probability that a large random matrix has no real eigenvalues? (2015). arXiv:1503.07926

42. Dyson, F.J.: A Brownian-motion model for the Eigenvalues of a random matrix. J. Math. Phys. 3, 1191–1198 (1962)

43. Majumdar, S.N., Nadal, C., Scardicchio, A., Vivo, P.: Index distribution of gaussian random matrices. Phys. Rev. Lett. 103(22), 220603 (2009)

44. Majumdar, S.N., Vivo, P.: Number of relevant directions in principal component analysis and wishart random matrices. Phys. Rev. Lett. 108(20), 200601 (2012)

45. Rogers, L., Shi, Z.: Interacting Brownian particles and the Wigner law. Probab. Theory Relat. Fields 95, 555–570 (1993)

46. Cepa, E., Lepingle, D.: Diffusing particles with electrostatic repulsion. Probab. Theory Relat. Fields 107, 429–449 (1997)

47. Anderson, W., Guionnet, A., Zeitouni, O.: An Introduction to Random Matrices. Cambridge University Press, Cambridge (2009)

48. Sandier, E., Serfaty, S.: 1d log gases and the renormalized energy: crystallization at vanishing temperature. Probab. Theory Relat Fields 162, 1–52 (2014)

49. Sandier, E., Serfaty, S.: 2d coulomb gases and the renormalized energy (2012). arXiv:1201.3503

50. Rougerie, N., Serfaty, S.: Higher dimensional coulomb gases and renormalized energy functionals (2013). arXiv:1307.2805

51. Allez, R., Touboul, J., Wainrib, G.: Index distribution of the Ginibre ensemble. J. Phys. A 47(4), 042001 (2014)

52. Braides, A.: Gamma-convergence for Beginners. Oxford University Press, Oxford (2002)

53. Armstrong, S.N., Serfaty, S., Zeitouni, O.: Remarks on a constrained optimization problem for the ginibre ensemble. Potential Anal. 41(3), 945–958 (2014)

54. Abrikosov, A.A.: On the magnetic properties of superconductors of the second type. Sov. Phys. JETP 5, 1174–1182 (1957)

55. Chafaï, D., Gozlan, N., Zitt, P.-A.: First order global asymptotics for confined particles with singular pair repulsion (2013). arXiv:1304.7569

56. del Molino, L.C.G., Pakdaman, K., Touboul, J.: The heterogeneous gas with singular interaction: generalized circular law and heterogeneous renormalized energy. J. Phys. A 48(4), 045208 (2015)

57. Vivo, P., Majumdar, S.N., Bohigas, O.: Large deviations and random matrices. Acta Phys. Pol. B 38(13), 4139 (2007)

## Acknowledgments

We thank an anonymous referee for his suggestions on the proof of Theorem 1.

## Author information

Authors

### Corresponding author

Correspondence to Luis Carlos García del Molino.

## Appendices

### Monte Carlo Algorithm for the Eigenvalues

An efficient method to approximate numerically the minimizer $$\mu _{\alpha }$$ and the probability distribution of the proportion of real eigenvalues is to use the Metropolis–Hastings Monte Carlo algorithm. This method consists in constructing an ergodic Markov chain whose stationary distribution is given by (1). Here, we evolve a n-particles system $$z_t$$, but in contrast to the log-gas, the dynamics is now discrete, and the transition probability is based on the pdf (1): a new configuration $$z^*$$ is drawn by modifying one of the eigenvalues at random and the Markov chain has a transition towards $$z^*$$ if $$Q^n(z^*)>Q^n(z_t)$$, and otherwise according to a Bernoulli variable of parameter $$\frac{Q^n(z^*)}{Q^n(z_t)}$$.

When conditioning on very rare events, (here for instance, a fixed number of real eigenvalues), cases satisfying the constraints have an extremely low probability of being explored, and more refined methods need to be developed in order to access these probabilities. In the present case, the problem is considerably simplified since we dispose of an explicit form of the distribution of the eigenvalues under our constraint. Indeed, the joint probability distribution of Ginibre matrices of size n constrained on having k real eigenvalues $$(\lambda _i\;;\; i=1\ldots k)$$ (and therefore $$l=(n-k)/2$$ pairs of complex eigenvalues $$(z_i, i=1\ldots n-k)$$) is given by:

\begin{aligned}&\mathbb {P}\big [\lambda _1\ldots \lambda _k, z_1,\ldots ,z_{n-k}\big ]=\tilde{C}_n \prod _{i>j}\vert \lambda _i-\lambda _j\vert \prod _{i>j} \vert z_i-z_j\vert \prod _{i,j}\vert \lambda _i-z_j\vert \\&\qquad \qquad \qquad \times \left( \prod _{i=1}^k\exp (-\lambda _i^2)\prod _{i=1}^{n-k} \exp (-z_i^2)\text {erfc}(\vert z_i-z_i^*\vert /\sqrt{2})\right) ^{1/2}, \end{aligned}

where the coefficient $$C_n$$ can be found in [14].

Classical Metropolis–Hastings algorithm with Gaussian transitions preserving the nature of the system therefore allow to access directly the distribution of eigenvalues and the probability p(nk) of the event considered.

## Rights and permissions

Reprints and permissions

del Molino, L.C.G., Pakdaman, K., Touboul, J. et al. The Real Ginibre Ensemble with $$k=O(n)$$ Real Eigenvalues. J Stat Phys 163, 303–323 (2016). https://doi.org/10.1007/s10955-016-1485-0