An observation about submatrices

Sourav Chatterjee (University of California at Berkeley)
Michel Ledoux (Institut de Mathematiques, Universite de Toulouse)

Abstract


Let $M$ be an arbitrary Hermitian matrix of order $n$, and $k$ be a positive integer less than $n$. We show that if $k$ is large, the distribution of eigenvalues on the real line is almost the same for almost all principal submatrices of $M$ of order $k$. The proof uses results about random walks on symmetric groups and concentration of measure. In a similar way, we also show that almost all $k \times n$ submatrices of $M$ have almost the same distribution of singular values.

Full Text: Download PDF | View PDF online (requires PDF plugin)

Pages: 495-500

Publication Date: November 5, 2009

DOI: 10.1214/ECP.v14-1504

References

  1. Bai, Z. D. Methodologies in spectral analysis of large-dimensional random matrices, a review.With comments by G. J. Rodgers and Jack W. Silverstein; and a rejoinder by the author. Statist. Sinica 9 (1999), no. 3, 611--677. MR1711663 (2000e:60044)
  2. Bobkov, Sergey G. Concentration of normalized sums and a central limit theorem for noncorrelated random variables. Ann. Probab. 32 (2004), no. 4, 2884--2907. MR2094433 (2005i:60041)
  3. Bobkov, Sergey G.; Tetali, Prasad. Modified logarithmic Sobolev inequalities in discrete settings. J. Theoret. Probab. 19 (2006), no. 2, 289--336. MR2283379 (2007m:60215)
  4. Diaconis, Persi; Shahshahani, Mehrdad. Generating a random permutation with random transpositions. Z. Wahrsch. Verw. Gebiete 57 (1981), no. 2, 159--179. MR0626813 (82h:60024)
  5. Ledoux, Michel. The concentration of measure phenomenon.Mathematical Surveys and Monographs, 89. American Mathematical Society, Providence, RI, 2001. x+181 pp. ISBN: 0-8218-2864-9 MR1849347 (2003k:28019)
  6. Rudelson, Mark; Vershynin, Roman. Sampling from large matrices: an approach through geometric functional analysis. J. ACM 54 (2007), no. 4, Art. 21, 19 pp. (electronic). MR2351844 (2009d:68177)


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.