Electronic Journal of Probability

Asymptotic freeness for rectangular random matrices and large deviations for sample covariance matrices with sub-Gaussian tails

Benjamin Groux

Full-text: Open access

Abstract

We establish a large deviation principle for the empirical spectral measure of a sample covariance matrix with sub-Gaussian entries, which extends Bordenave and Caputo’s result for Wigner matrices having the same type of entries [7]. To this aim, we need to establish an asymptotic freeness result for rectangular free convolution, more precisely, we give a bound in the subordination formula for information-plus-noise matrices.

Article information

Source
Electron. J. Probab., Volume 22 (2017), paper no. 53, 40 pp.

Dates
Received: 15 May 2015
Accepted: 9 September 2016
First available in Project Euclid: 21 June 2017

Permanent link to this document
https://projecteuclid.org/euclid.ejp/1498010466

Digital Object Identifier
doi:10.1214/17-EJP4326

Mathematical Reviews number (MathSciNet)
MR3666016

Zentralblatt MATH identifier
1380.60038

Subjects
Primary: 60B20: Random matrices (probabilistic aspects; for algebraic aspects see 15B52) 46L54: Free probability and free operator algebras 60F10: Large deviations

Keywords
random matrices large deviations free convolution subordination property spectral measure Stieltjes transform information-plus-noise model

Rights
Creative Commons Attribution 4.0 International License.

Citation

Groux, Benjamin. Asymptotic freeness for rectangular random matrices and large deviations for sample covariance matrices with sub-Gaussian tails. Electron. J. Probab. 22 (2017), paper no. 53, 40 pp. doi:10.1214/17-EJP4326. https://projecteuclid.org/euclid.ejp/1498010466


Export citation

References

  • [1] G. W. Anderson, A. Guionnet, and O. Zeitouni. An introduction to random matrices, volume 118 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2010.
  • [2] F. Augeri. Large deviations principle for the largest eigenvalue of Wigner matrices without Gaussian tails. arXiv:1502.07983v1, 2015.
  • [3] Z. Bai and J. W. Silverstein. Spectral analysis of large dimensional random matrices. Springer Series in Statistics. Springer, New York, second edition, 2010.
  • [4] G. Ben Arous and A. Guionnet. Large deviations for Wigner’s law and Voiculescu’s non-commutative entropy. Probab. Theory Related Fields, 108(4):517–542, 1997.
  • [5] F. Benaych-Georges. Rectangular random matrices, related convolution. Probab. Theory Related Fields, 144(3–4):471–515, 2009.
  • [6] P. Biane. On the free convolution with a semi-circular distribution. Indiana Univ. Math. J., 46(3):705–718, 1997.
  • [7] C. Bordenave and P. Caputo. A large deviation principle for Wigner matrices without Gaussian tails. Ann. Probab., 42(6):2454–2496, 2014.
  • [8] M. Capitaine, C. Donati-Martin, D. Féral, and M. Février. Free convolution with a semicircular distribution and eigenvalues of spiked deformations of Wigner matrices. Electron. J. Probab., 16(64):1750–1792, 2011.
  • [9] A. Dembo and O. Zeitouni. Large deviations techniques and applications, volume 38 of Applications of Mathematics (New York). Springer-Verlag, New York, second edition, 1998.
  • [10] R. B. Dozier and J. W. Silverstein. On the empirical distribution of eigenvalues of large dimensional information-plus-noise-type matrices. J. Multivariate Anal., 98(4):678–694, 2007.
  • [11] J. Dumont, W. Hachem, S. Lasaulce, P. Loubaton, and J. Najim. On the capacity achieving covariance matrix for Rician MIMO channels: an asymptotic approach. IEEE Trans. Inform. Theory, 56(3):1048–1069, 2010.
  • [12] P. Eichelsbacher, J. Sommerauer, and M. Stolz. Large deviations for disordered bosons and multiple orthogonal polynomial ensembles. J. Math. Phys., 52(7):073510, 16, 2011.
  • [13] A. Fey, R. van der Hofstad, and M. J. Klok. Large deviations for eigenvalues of sample covariance matrices, with applications to mobile communication systems. Adv. in Appl. Probab., 40(4):1048–1071, 2008.
  • [14] B. Groux. Grandes déviations de matrices aléatoires et Équation de Fokker-Planck libre. PhD thesis, Université Paris-Saclay, 2016.
  • [15] A. Guionnet and O. Zeitouni. Concentration of the spectral measure for large matrices. Electron. Comm. Probab., 5:119–136 (electronic), 2000.
  • [16] F. Hiai and D. Petz. Eigenvalue density of the Wishart matrix and large deviations. Infin. Dimens. Anal. Quantum Probab. Relat. Top., 1(4):633–646, 1998.
  • [17] F. Hiai and D. Petz. The semicircle law, free random variables and entropy, volume 77 of Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2000.
  • [18] V. A. Marčenko and L. A. Pastur. Distribution of eigenvalues in certain sets of random matrices. Mat. Sb. (N.S.), 72(114):507–536, 1967.
  • [19] L. Pastur and M. Shcherbina. Eigenvalue distribution of large random matrices, volume 171 of Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2011.
  • [20] P. Vallet, P. Loubaton, and X. Mestre. Improved subspace estimation for multivariate observations of high dimension: the deterministic signals case. IEEE Trans. Inform. Theory, 58(2):1043–1068, 2012.
  • [21] E. P. Wigner. On the distribution of the roots of certain symmetric matrices. Ann. of Math. (2), 67:325–327, 1958.
  • [22] X. Zhan. Matrix inequalities, volume 1790 of Lecture Notes in Mathematics. Springer-Verlag, Berlin, 2002.