### Abstract

Assume a finite set of complex random variables form a determinantal point process; we obtain a theorem on the limit of the empirical distribution of these random variables. The result is applied to two types of n-by-n random matrices as n goes to infinity. The first one is the product of m i.i.d. (complex) Ginibre ensembles, and the second one is the product of truncations of m independent Haar unitary matrices with sizes n _{j} × n _{j} for 1 ≤ j≤ m. Assuming m depends on n, by using the special structures of the eigenvalues we developed, explicit limits of spectral distributions are obtained regardless of the speed of m compared to n. For the product of m Ginibre ensembles, as m is fixed, the limiting distribution is known by various authors, e.g., Götze and Tikhomirov (On the asymptotic spectrum of products of independent random matrices, 2010. http://arxiv.org/pdf/1012.2710v3.pdf), Bordenave (Electron Commun Probab 16:104–113, 2011), O’Rourke and Soshnikov (Electron J Probab 16(81):2219–2245, 2011) and O’Rourke et al. (J Stat Phys 160(1):89–119, 2015). Our results hold for any m≥ 1 which may depend on n. For the product of truncations of Haar-invariant unitary matrices, we show a rich feature of the limiting distribution as n _{j} / n’s vary. In addition, some general results on arbitrary rotation-invariant determinantal point processes are also derived. In particular, we obtain an inequality for the fourth moment of linear statistics of complex random variables forming a determinantal point process. This inequality is known for the complex Ginibre ensemble only (Hwang in Random matrices and their applications (Brunswick, Maine, 1984), Contemporary Mathematics, American Mathematics Society, Providence, vol 50, pp 145–152, 1986). Our method is the determinantal point process rather than the contour integral by Hwang.

Original language | English (US) |
---|---|

Pages (from-to) | 353-394 |

Number of pages | 42 |

Journal | Journal of Theoretical Probability |

Volume | 32 |

Issue number | 1 |

DOIs | |

State | Published - Mar 15 2019 |

### Fingerprint

### Keywords

- Determinantal point process
- Eigenvalue
- Empirical distribution
- Non-symmetric random matrix

### Cite this

**Empirical Distributions of Eigenvalues of Product Ensembles.** / Jiang, Tiefeng; Qi, Yongcheng.

Research output: Contribution to journal › Article

*Journal of Theoretical Probability*, vol. 32, no. 1, pp. 353-394. https://doi.org/10.1007/s10959-017-0799-4

}

TY - JOUR

T1 - Empirical Distributions of Eigenvalues of Product Ensembles

AU - Jiang, Tiefeng

AU - Qi, Yongcheng

PY - 2019/3/15

Y1 - 2019/3/15

N2 - Assume a finite set of complex random variables form a determinantal point process; we obtain a theorem on the limit of the empirical distribution of these random variables. The result is applied to two types of n-by-n random matrices as n goes to infinity. The first one is the product of m i.i.d. (complex) Ginibre ensembles, and the second one is the product of truncations of m independent Haar unitary matrices with sizes n j × n j for 1 ≤ j≤ m. Assuming m depends on n, by using the special structures of the eigenvalues we developed, explicit limits of spectral distributions are obtained regardless of the speed of m compared to n. For the product of m Ginibre ensembles, as m is fixed, the limiting distribution is known by various authors, e.g., Götze and Tikhomirov (On the asymptotic spectrum of products of independent random matrices, 2010. http://arxiv.org/pdf/1012.2710v3.pdf), Bordenave (Electron Commun Probab 16:104–113, 2011), O’Rourke and Soshnikov (Electron J Probab 16(81):2219–2245, 2011) and O’Rourke et al. (J Stat Phys 160(1):89–119, 2015). Our results hold for any m≥ 1 which may depend on n. For the product of truncations of Haar-invariant unitary matrices, we show a rich feature of the limiting distribution as n j / n’s vary. In addition, some general results on arbitrary rotation-invariant determinantal point processes are also derived. In particular, we obtain an inequality for the fourth moment of linear statistics of complex random variables forming a determinantal point process. This inequality is known for the complex Ginibre ensemble only (Hwang in Random matrices and their applications (Brunswick, Maine, 1984), Contemporary Mathematics, American Mathematics Society, Providence, vol 50, pp 145–152, 1986). Our method is the determinantal point process rather than the contour integral by Hwang.

AB - Assume a finite set of complex random variables form a determinantal point process; we obtain a theorem on the limit of the empirical distribution of these random variables. The result is applied to two types of n-by-n random matrices as n goes to infinity. The first one is the product of m i.i.d. (complex) Ginibre ensembles, and the second one is the product of truncations of m independent Haar unitary matrices with sizes n j × n j for 1 ≤ j≤ m. Assuming m depends on n, by using the special structures of the eigenvalues we developed, explicit limits of spectral distributions are obtained regardless of the speed of m compared to n. For the product of m Ginibre ensembles, as m is fixed, the limiting distribution is known by various authors, e.g., Götze and Tikhomirov (On the asymptotic spectrum of products of independent random matrices, 2010. http://arxiv.org/pdf/1012.2710v3.pdf), Bordenave (Electron Commun Probab 16:104–113, 2011), O’Rourke and Soshnikov (Electron J Probab 16(81):2219–2245, 2011) and O’Rourke et al. (J Stat Phys 160(1):89–119, 2015). Our results hold for any m≥ 1 which may depend on n. For the product of truncations of Haar-invariant unitary matrices, we show a rich feature of the limiting distribution as n j / n’s vary. In addition, some general results on arbitrary rotation-invariant determinantal point processes are also derived. In particular, we obtain an inequality for the fourth moment of linear statistics of complex random variables forming a determinantal point process. This inequality is known for the complex Ginibre ensemble only (Hwang in Random matrices and their applications (Brunswick, Maine, 1984), Contemporary Mathematics, American Mathematics Society, Providence, vol 50, pp 145–152, 1986). Our method is the determinantal point process rather than the contour integral by Hwang.

KW - Determinantal point process

KW - Eigenvalue

KW - Empirical distribution

KW - Non-symmetric random matrix

UR - http://www.scopus.com/inward/record.url?scp=85037606637&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037606637&partnerID=8YFLogxK

U2 - 10.1007/s10959-017-0799-4

DO - 10.1007/s10959-017-0799-4

M3 - Article

AN - SCOPUS:85037606637

VL - 32

SP - 353

EP - 394

JO - Journal of Theoretical Probability

JF - Journal of Theoretical Probability

SN - 0894-9840

IS - 1

ER -